Air Force Institute of Technology Air Force Institute of Technology
AFIT Scholar AFIT Scholar
Theses and Dissertations Student Graduate Works
3-17-2008
Computational Design of Upperstage Chamber, Aerospike, & Computational Design of Upperstage Chamber, Aerospike, &
Cooling Jacket for Dual-Expander Rocket Engine Cooling Jacket for Dual-Expander Rocket Engine
David F. Martin II
Follow this and additional works at: https://scholar.afit.edu/etd
Part of the Propulsion and Power Commons
Recommended Citation Recommended Citation Martin, David F. II, "Computational Design of Upperstage Chamber, Aerospike, & Cooling Jacket for Dual-Expander Rocket Engine" (2008). Theses and Dissertations. 2686. https://scholar.afit.edu/etd/2686
This Thesis is brought to you for free and open access by the Student Graduate Works at AFIT Scholar. It has been accepted for inclusion in Theses and Dissertations by an authorized administrator of AFIT Scholar. For more information, please contact [email protected].
‘t have to p[
COMPUTATIONAL DESIGN OF UPPERSTAGE CHAMBER, AEROSPIKE, & COOLING JACKET FOR DUAL-EXPANDER ROCKET ENGINE
THESIS
David F. Martin II, 2Lt, USAF
AFIT/GAE/ENY/08-M20
DEPARTMENT OF THE AIR FORCE AIR UNIVERSITY
AIR FORCE INSTITUTE OF TECHNOLOGY
Wright-Patterson Air Force Base, Ohio
APPROVED FOR PUBLIC RELEASE; DISTRIBUTION UNLIMITED
The views expressed in this thesis are those of the author and do not reflect the official
policy or position of the United States Air Force, Department of Defense, or the U.S.
Government.
AFIT/GAE/ENY/08-M20
COMPUTATIONAL DESIGN OF UPPERSTAGE CHAMBER, AEROSPIKE, & COOLING JACKET OF DUAL-EXPANDER ROCKET ENGINE
THESIS
Presented to the Faculty
Department of ENY
Graduate School of Engineering and Management
Air Force Institute of Technology
Air University
Air Education and Training Command
In Partial Fulfillment of the Requirements for the
Degree of Master of Science in Aerospace Engineering
David F. Martin II, BS
2Lt, USAF
March 2008
APPROVED FOR PUBLIC RELEASE; DISTRIBUTION UNLIMITED
AFIT/GAE/ENY/08-M20
COMPUTATIONAL DESIGN OF UPPERSTAGE CHAMBER, AEROSPIKE, & COOLING JACKET OF DUAL-EXPANDER ROCKET ENGINE
David F. Martin II, BS
2Lt, USAF
Approved: ________//signed//______________________ 17 Mar 08 Richard D. Branam, Maj, USAF (Chairman) Date ________//signed//_______________________ 17 Mar 08 Raymond C. Maple, Lt Col, USAF (Member) Date
________//signed//_______________________ 17 Mar 08 Richard E. Huffman, Maj, USAF (Member) Date
v
Acknowledgments
I would like to express my sincere appreciation to my faculty advisor, Maj Richard
Branam, for his guidance and support throughout the course of this thesis effort. The
insight and experience was certainly appreciated. I would also like to thank my sponsor,
Mr. Michael Huggins, from the AFRL at Edwards for the support. I want to thank Capt
Michael Arguello and Capt William Strain, my fellow workers, who managed to endure
on this project despite themselves. I would also like to thank Thomas Lavelle at NASA
Glenn for his help with NPSS. Without him the model would never have been
completed. I thank Timothy O’Brien and Ronald Springer of Aerojet for their TDK and
NPSS code and their help. Finally, a big thanks goes out to my family and friends who
supported me through thick and thin.
David F. Martin II
vi
Table of Contents
Page
Table of Contents ............................................................................................................... vi
List of Figures ......................................................................................................................x
List of Tables .................................................................................................................... xii
List of Symbols ................................................................................................................ xiii
Abstract ............................................................................................................................ xvi
1. Introduction ................................................................................................................17
1.1. Motivation ......................................................................................................17
1.2. Objectives ......................................................................................................18
1.2.1. The DEAN .............................................................................................. 19
1.2.2. Thrust Chamber...................................................................................... 20
1.2.3. Nozzle ..................................................................................................... 20
1.2.4. Cooling Jacket ........................................................................................ 21
1.3. Preview ..........................................................................................................21
2. Background ................................................................................................................22
2.1. Orbit Transfer Engines and Mission Requirements .......................................22
2.1.1. Baseline Engine ...................................................................................... 22
2.1.2. Typical Mission Requirements ............................................................... 24
2.1.3. Expander Cycles ..................................................................................... 25
2.2. Thrust Chamber .............................................................................................26
2.2.1. Rocket Engine Performance ................................................................... 26
2.2.2. Dimension of the Thrust Chamber ......................................................... 28
2.2.3. Injectors .................................................................................................. 29
vii
2.3. Nozzle ............................................................................................................33
2.3.1. Aerospike ................................................................................................ 34
2.4. Cooling Jacket ................................................................................................40
2.4.1. Heat Transfer ......................................................................................... 40
2.4.2. Cooling Techniques ................................................................................ 43
2.4.3. Aerospike Cooling .................................................................................. 45
2.4.4. Material .................................................................................................. 47
2.4.5. Aspect-Ratio ........................................................................................... 48
3. Methodology ..............................................................................................................52
3.1. General Sizing ................................................................................................52
3.2. Thrust Chamber .............................................................................................54
3.2.1. Chamber Controls .................................................................................. 54
3.2.2. Chamber Process Variables ................................................................... 54
3.2.3. Chamber Outputs ................................................................................... 54
3.3. Nozzle ............................................................................................................55
3.3.1. Nozzle Controls ...................................................................................... 55
3.3.2. Nozzle Process Variables ....................................................................... 55
3.3.3. Nozzle Outputs ........................................................................................ 55
3.4. Cooling Jacket ................................................................................................55
3.4.1. Cooling Jacket Controls ......................................................................... 56
3.4.2. Cooling Jacket Process Variables .......................................................... 56
3.4.3. Cooling Jacket Outputs .......................................................................... 56
viii
3.5. NPSS ..............................................................................................................56
3.6. TDK ...............................................................................................................60
3.7. Sensitivity Analysis .......................................................................................61
3.8. Beginning Frame Work ..................................................................................63
3.8.1. Cooling Channel Design ........................................................................ 64
3.8.2. Material Choice ...................................................................................... 65
4. Analysis and Results ..................................................................................................68
4.1. Chapter Overview ..........................................................................................68
4.2. Results of Simulation Scenarios ....................................................................68
4.2.1. Hydrogen Cooling Jacket Results .......................................................... 69
4.2.2. Oxygen Cooling Jacket Results .............................................................. 73
4.2.3. Chamber/Nozzle Results ......................................................................... 76
4.2.4. TDK Nozzle ............................................................................................ 77
4.3. Investigative Objectives .................................................................................81
4.4. Summary ........................................................................................................82
5. Conclusions and Recommendations ...........................................................................84
5.1. Chapter Overview ..........................................................................................84
5.2. Conclusions of Research ................................................................................84
5.3. Significance of Research ................................................................................84
5.4. Recommendations for Future Research .........................................................85
5.5. Summary ........................................................................................................86
Appendix A : Lessons Learned ..........................................................................................87
A.1 NPSS Lessons Learned ........................................................................................87
ix
A.2 TDK Lessons Learned .........................................................................................88
Appendix B : NPSS Code ..................................................................................................89
B.1 Final Model Code ................................................................................................89
B.2 Included Cooling Volume Element ...................................................................105
B.3 Included Pump Element .....................................................................................114
B.4 Included Turbine Element .................................................................................118
Appendix C : TDK Code .................................................................................................123
C.1 TDK Input for DEAN Upper Stage Engine .......................................................123
C.2 TDK Performance Summary for the DEAN ......................................................124
Appendix D : Isp Calculation ............................................................................................127
Appendix E : Wall Temperatures ...................................................................................128
References ........................................................................................................................131
Vita. ..................................................................................................................................134
x
List of Figures
Figure Page
Figure 1. The Dean Schematic .......................................................................................... 19
Figure 2. RL10B-2(with permission from Pratt & Whitney)3 .......................................... 23
Figure 3. RL10B-2 Cycle Schematic (with permission from Pratt & Whitney)3 ............. 25
Figure 4. Shear Coaxial Gas/Liquid Injector .................................................................... 30
Figure 5. Pintle Injector .................................................................................................... 32
Figure 6. Aerospike ........................................................................................................... 33
Figure 7. Flow Phenomena of Plug Nozzle ...................................................................... 35
Figure 8. Flow Phenomena of Truncated Plug Nozzle ..................................................... 36
Figure 9. Channel Aspect Ratio ....................................................................................... 49
Figure 10. NPSS Model Schematic .................................................................................. 57
Figure 11. NPSS Cooling Jacket ....................................................................................... 59
Figure 12. Sensitivity Model ............................................................................................. 61
Figure 13. Initial Contour ................................................................................................. 63
Figure 14. Cooling Channel Cross-Section ...................................................................... 64
Figure 15. Chamber/Nozzle Contour (all dimension in inches) ....................................... 68
Figure 16. Hydrogen Wall Temperatures ......................................................................... 72
Figure 17. Oxygen Wall Temperatures ............................................................................. 75
Figure 18. Non-dimensional Nozzle Contour with Temperature Profile ......................... 78
Figure 19. Truncated Nozzle ............................................................................................. 79
Figure 20. Nozzle Contour ................................................................................................ 80
xi
Figure 21. The DEAN ....................................................................................................... 81
xii
List of Tables
Table Page
Table 1. Average Price per Pound of Launch Vehicles .................................................... 18
Table 2. Influence on c*7 .................................................................................................. 27
Table 3. Stepped Channel vs. Constant Cross-Section Channel42 .................................... 51
Table 4. Chosen Parameters .............................................................................................. 52
Table 5. Hydrogen Sensitivity Analysis ........................................................................... 62
Table 6. Oxygen Sensitivity Analysis ............................................................................... 62
Table 7. Material Thermal Properties19 ............................................................................ 65
Table 8. Structural Materials ............................................................................................. 66
Table 9. Oxygen Compatibility ......................................................................................... 67
Table 10. Hydrogen Channel Dimensions ........................................................................ 70
Table 11. Hydrogen Mach Numbers ................................................................................. 70
Table 12. Hydrogen Flow Temperature ............................................................................ 73
Table 13. Oxygen Channel Dimensions ........................................................................... 74
Table 14. Oxygen Mach Numbers .................................................................................... 74
Table 15. Oxygen Flow Temperature ............................................................................... 76
Table 16. Chamber Performance ...................................................................................... 77
Table 17. TDK Properties ................................................................................................. 78
Table 18. Truncation of Nozzle ........................................................................................ 79
Table 19 Hydrogen Wall Materials ................................................................................ 128
Table 20 Oxygen Wall Materials .................................................................................... 129
xiii
List of Symbols
Symbol
a speed of sound, channel half spacing Ac combustion chamber area, cold side area Ae exit area AH hot side area At throat area AR aspect ratio A* characteristic area BF burn factor Btu British thermal unit c* characteristic velocity C constant Cg throat region correlation coefficient CT thrust coefficient cal calorie CFD computational fluid dynamics CH4 methane DEAN Dual Expander Aerospike Nozzle engine dT/dx derivative of the temperature with respect to position EELV Evolved Expendable Launch Vehicle FESTIP Future European Space Transportation Investigations Program Fg thrust ft foot Ftu ultimate tensile strength GEO geostationary orbit gmole 6.02252*1023 molecules g0 acceleration due to gravity GTO geo transfer orbit GUI graphical user interface h channel height , convective heat transfer coefficient hc coolant convective heat transfer coefficient hH hot side convective heat transfer coefficient ΔH0
f heat of oxidation IHPRPT Integrated High Payoff Rocket Propulsion Technology initiative in inch Isp specific impulse k thermal conductivity K Kelvin Kg kilograms L length L* characteristic length
xiv
lbf ponds force lbm pounds mass LEO low earth orbit LH2 liquid hydrogen LO2, LOX liquid oxygen m meter m mass flow rate M molecular weight M Mach number Mb burnout mass Me exit Mach number Mo initial mass min minute mm millimeter N Newton’s NASA National Air and Space Association NIST National Institute of Standards and Technology NPSS TM Numerical Propulsion System Simulation Nu Nusselt number O/F oxidizer-to-fuel ratio P0 total chamber pressure Pa Pascal Pa ambient Pb burst pressure Pc chamber pressure Pe exit pressure PR pressure ratio Pr* reference Prandtl number psi pounds per square inch q’’ heat flux qx heat rate x-direction Q parameter r radius R Rankine R* throat radius R Universal gas constant rc chamber radius Re*d reference Reynolds’s number RP1 hydrocarbon fuel s, sec seconds SSME space shuttle main engine St* reference Stewart number t thickness TC cold side temperature
xv
TDK 04TM Two-dimensional kinematics TH hot side temperature T0 total temperature in the chamber Ts temperature of the surface tw wall thickness Twall temperature of the wall T∞ temperature of the fluid Δu change in velocity VCCW vortex combustion cold-wall chamber Vc chamber volume w channel half width W watts X nozzle length α thermal diffusivity γ ratio of specific heat Δ change in ε expansion ratio λ nozzle efficiency
xvi
AFIT/GAE/ENY/08-M20
Abstract
To increase the performance of the current US satellite launch capability, new
rocket designs must be undertaken. One concept that has been around since the 50s but
yet to be utilized on a launch platform is the aerospike, or plug nozzle. The aerospike
nozzle concept demonstrates globally better performance compared to a conventional bell
nozzle, since the expansion of the jet is not bounded by a wall and therefore can adjust to
the environment by changing the outer jet boundary. A dual-expander aerospike nozzle
(DEAN) rocket concept would exceed the Integrated High Payoff Rocket Propulsion
Technology initiative (IHPRPT) phase three goals. This document covers the design of
the chamber and nozzle of the DEAN. The validation of the design of the DEAN are
based on the model in Numerical Propulsion System Simulation (NPSS TM), added with
the nozzle design from Two-Dimensional Kinematics (TDK 04TM). The result is a rocket
engine that produces 57,231 lbf (254.5 kN) of thrust at an Isp of 472 s. Additionally, the
oxygen wall is made of silicon carbide, with a melting point of 5580 R (3100 K), and has
a maximum temperature at the throat of 1625 R (902 K). The hydrogen side is made of
copper, with a melting point of 2444 R (1358 K), and has a maximum wall temperature
of 1224 R (680 K) at the throat. Based on these result, future investigation into this
design is merited since it has the potential to save $19 million in the cost to launch a
satellite. NPSS proved to be a powerful tool in the development of rocket engines. TDK,
however, was left wanting in the area of aerospike design.
17
COMPUTATIONAL DESIGN OF UPPERSTAGE CHAMBER, AEROSPIKE, & COOLING JACKET OF DUAL-EXPANDER ROCKET ENGINE
1. Introduction
Engineers live to design at the edge of what is possible. Since the beginning of
rocketry, the bell nozzle has dominated the rocket motor design. Now the time is right to
revive an old concept and couple it with new technology to break the bell’s dominance.
The concept is the aerospike nozzle and the dual-expander aerospike nozzle, or DEAN, is
the design that will usher in an age of improved performance and cost in the space launch
arena.
1.1. Motivation
The Integrated High Payoff Rocket Propulsion Technology initiative (IHPRPT)1
is a joint government and industry effort focused on developing affordable technologies
for reach capability, sustainable strategic missiles, long life or increased maneuverability
spacecraft capability and high performance tactical missile capability. The objectives for
the boost and orbit transfer part is to increase the specific impulse (Isp), increase the
thrust-to-weight ratio or the mass fraction, reduce the failure rate, and increase the
reusability. All of this will lead to the end goal of a reduction in the cost of launching
satellites. The goals of phase III are to increase the Isp by 26 seconds and improve thrust-
to-weight by 100%. The result of the program should be an increase in payload of 22%
and 95% and a reduction in cost of 33% and 82% for expendable and reusable launch
vehicles, respectively.
One way to determine the cost of a launch is per pound. Table 1 shows the
estimated average cost per pound.
18
Table 1. Average Price per Pound of Launch Vehicles2
Vehicle Class
LEO GTO
Western Non-Western Western
Non-Western
Small $8,445 $3,208 $18,841 N/A Medium $4,994 $2,404 $12,133 $9,843 Heavy $4,440 $1,946 $17,032 $6,967
In a simple analysis detailed in Appendix D, for the same change in velocity (Δu)
a one-second increase in Isp results in a savings of 134.5 lbm (61 kg). Based on the
numbers from Table 1 for medium western vehicles, this results in a savings of as much
as $671,693 a launch, or larger and more capable satellites. Moreover, that only
represents a single second increase in Isp so the actual savings could be dramatically
greater.
1.2. Objectives
To meet the goals of IHPRPT, a Dual Expander Aerospike Nozzle engine
(DEAN) is being designed to provide 50,000 lbf (222.4 kN) of thrust with an Isp of 464 s.
This research effort focuses on the development of a thrust chamber, nozzle, and cooling
jacket in an attempt to satisfy the following three goals.
1. Determine feasibility of meeting the IHPRPT Phase III orbit transfer vehicle
goals with the DEAN concept
2. Implement and improve upon a design process focused on the energy
conversion section of a rocket engine (combustion chamber, nozzle)
19
3. Perform detailed design analysis of the energy transfer components (cooling
jackets) making the DEAN possible
1.2.1. The DEAN
The DEAN will utilize liquid hydrogen and oxygen as the fuel and oxidizer. Each
will operate in their own expander cycle powered by the heat from the thrust chamber
and nozzle. The DEAN is also designed with an aerospike or plug nozzle. The DEAN
has the potential to dramatically reduce the cost of launching a satellite. Figure 1 show a
schematic of the Dean.
Figure 1. The Dean Schematic
Figure 1 shows that each of the fluid flows is contain their own turbo-machinery
and cool a separate wall (oxygen cools the outer-wall and hydrogen cools the inner-wall).
Four system design choices make the DEAN a revolutionary engine. First, the Dean
20
utilizes the aerospike nozzle that will dramatically reduce the weight and improved
performance over a bell nozzle. Since the fuel and oxidizer are separated until injection
into the chamber there is no need for inter propellant seals thus reducing a critical failure
mode increasing reliability. The split flow on the fuel side reduces the required
horsepower that increases the life of the turbo-machinery. Lastly, the design has the
ability to be throttled.
In order to analyze the design of the DEAN, two computational programs will be
used. Two-dimensional kinematics (TDK 04TM) will be used to design the contour of the
nozzle. Numerical Propulsion System Simulation (NPSS TM) will be used to asses the
performance of all the components of the DEAN. The thrust chamber, nozzle, and
cooling jacket are the focused of this document.
1.2.2. Thrust Chamber
The thrust chamber embodies the essence of rocket propulsion: the acceleration
of matter and the reaction imparting propulsive force to the vehicle. The aim is to
achieve a device of maximum performance, stability, durability while minimizing the
size, weight, and cost. In this report, the thrust chamber consists of the combustion
chamber and the injector. The combustion chamber provides a volume for proper mixing
of the propellants and length for complete combustion. The injector distributes the
prescribed propellant mass flows to the chamber. The key variables for the thrust
chamber are the geometry, pressure, and temperature of the chamber.
1.2.3. Nozzle
The nozzle is directly connected to the combustion chamber. The nozzle converts
the enthalpy of the hot combusted gases into kinetic energy and produces the thrust of the
21
engine. The key variables affected by the design of the nozzle are thrust, Isp, nozzle
length, and expansion ratio (ε). Maximizing the Isp is beneficial to any rocket design as
shown earlier. In the design of the nozzle, the ideal length is often quite long, adding
weight to the engine. Therefore, a design consideration is to minimizing the length of the
nozzle while still maintaining performance near the ideal case.
1.2.4. Cooling Jacket
The combustion temperatures in the thrust chamber are extremely high.
Additionally, the heat-transfer rates from the combusted gases to the wall are high.
Consequently, the cooling jacket requires major design consideration. Not only does the
jacket need to keep the walls cool enough to maintain their structural integrity, adequate
heat must be transferred to the cooling fluids to power the turbines. The cooling jacket
will be analyzed based on the temperature of the chamber wall and the temperature
change in the cooling fluid.
1.3. Preview
The remainder of this report begins with review of the subjects relevant to the
thrust chamber, nozzle, and cooling jacket. The methodology employed to conduct the
work within this report is presented next. The results obtained from the methodology
described is presented and discussed. Finally, the conclusions inferred based on the
results of this report are stated and any recommendations based on the work done are
declared.
22
2. Background
A substantial amount of work has already been completed covering the wide
range of topics relating to the design in this thesis. The purpose of this Chapter is to
present the prior knowledge that exists which can set benchmarks and help in the
execution of the design of the DEAN. Ideas and benchmarks from these works were
considered in the design of the three main elements in this thesis.
2.1. Orbit Transfer Engines and Mission Requirements
2.1.1. Baseline Engine
The DEAN was primarily designed to replace the Pratt and Whitney RL10 rocket
engine. The original RL10 was designed in 1959 as an upper-stage liquid-oxygen liquid-
hydrogen expander cycle rocket engine. The most current inceptions of the RL10 is the
RL10B-2. The RL10B-2 features the world’s largest carbon-carbon extendible nozzle.
This high-expansion ratio nozzle enables the RL10B-2 to achieve a remarkable 465.5 sec
of specific impulse and lift payloads of up to 30,000 lbm.3 Figure 2 shows an image of
the RL10B-2.
23
Figure 2. RL10B-2(with permission from Pratt & Whitney)Error! Bookmark not defined.
In Figure 2, the RL10B-2 is shown inside the expandable skirt. The RL10B-2
currently powers the upper stage of the medium and heavy-lift versions of Boeing’s Delta
IV for Evolved Expendable Launch Vehicle (EELV), government and commercial
missions.
The RL10 is also used on the Centaur upper stage. With the Titan IVB and the
Centaur upper-stage, they are able to insert payloads greater than 12,700 pounds directly
into geosynchronous orbit. The high-energy Centaur upper stage has evolved to become a
very versatile vehicle. Performing a three-burn mission, the Centaur achieves parking
24
orbit with the first burn, boosts itself and the satellite to a highly elliptical orbit with
second burn, and circularizes the orbit at geosynchronous altitude with the third burn4.
The Centaur propulsion system uses 2 RL10A-3-3A Pratt & Whitney engines,
Each engine produces 16,500 lbf of thrust, a 444.4 sec nominal Isp at 5.0:1 mixture ratio
and an area ratio of 61:1.4 This series of engines has been used successfully since 1963
on the Saturn and Atlas/Centaur vehicles. The RL10A-3-3A uses an expander cycle,
where all of the LH2 is burned in the combustion chamber, except for a small amount
used for autogenously pressurization and pump bearing cooling/gear box pressurization.4
The turbine working fluid is the supercritical hydrogen heated in the regeneratively
cooled thrust chamber.
2.1.2. Typical Mission Requirements
One of the simplest and most common methods of putting a spacecraft into
geostationary orbit involves three steps: launch to low earth (or parking) orbit with
chemical propulsion; erect a GTO (geostationary transfer orbit) with an additional
chemical stage; and perform a simultaneous circularization-and-plane-change maneuver
at the apogee of the GTO.5
With an Atlas/Centaur class launch vehicle, the conventional path to the
geostationary orbit (GEO) is a two-burn Centaur stage. The first burn establishes a
slightly elliptical parking orbit and the second burn places the satellite in a geostationary
transfer orbit.5
25
2.1.3. Expander Cycles
In general there are three classic engine system configurations; gas-generator
cycle, expander cycle, and staged combustion cycle. The DEAN utilizes the expander
cycle. The expander cycle places the turbine inline with the thrust chamber, exhausting
directly into the chamber.8 Figure 3 shows a schematic of the expander cycles of the
RL10B-2.
Figure 3. RL10B-2 Cycle Schematic (with permission from Pratt & Whitney)Error!
Bookmark not defined.
Like in Figure 3, most expander cycles use the fuel heated through cooling of the
chamber wall as the working fluid for the turbine.
The turbine powers the pumps that allow the propellants to be stored at lower
pressures, their by reducing the structural weight of the tanks. The limiting factor to the
Chamber
Nozzle
LOX Pump LH2 Pump
LH2 Turbine
Cooling Jacket
26
performance of an expander cycle is the turbine inlet temperature, that in turn limits the
attainable chamber pressure. Consequently, the expander engine is primarily used as a
space engine where it can exhaust to a vacuum and can have a very high nozzle area ratio
even though it has a lower chamber pressure.
2.2. Thrust Chamber
2.2.1. Rocket Engine Performance
A common performance parameter used to define a rocket engine is the specific
impulse (Isp). The Isp compares the thrust of the engine to the propellant mass flow rate.
Equation 1 show the Isp as defined by Humble, Henry & Larson:6
0gmFI sp = (1)
In Equation 1, F is the thrust, m is the propellant mass flow rate, and g0 is the
acceleration due to gravity. The units for Isp is seconds for both English or SI units.
In order to describe the performance of each component of the thrust chamber,
two coefficients are defined. For the combustion chamber, the characteristic velocity (c*)
characterizes the influence of propellant choice through absolute maximum temperature
achievable. For the nozzle, the thrust coefficient (CT) provides the conversion of the
potential energy to kinetic energy as well as the efficiency of the nozzle expansion.
Equations 2 and Equation 3 for c* and CT comes from Hill & Peterson:7
27
MTR
c 01
1
211*
−+
⎟⎠⎞
⎜⎝⎛ +
=γ
γγ
γ (2)
*1
12
12
0
1
0
11
2
AA
ppp
pp
C eaeeT
−+
⎥⎥
⎦
⎤
⎢⎢
⎣
⎡
⎟⎟⎠
⎞⎜⎜⎝
⎛−⎟⎟
⎠
⎞⎜⎜⎝
⎛+−
=
−−
+γ
γγ
γ
γγγ (3)
In Equation 2, γ is the ratio of specific heats, T0 is the total temperature in the
chamber, R is the universal gas constant, and M is the molecular weight of the
propellant. In Equation 3, pe, pa, and p0 are the exit, ambient, and total chamber pressure
respectively, Ae is the exit area and A* is a characteristic area for a Mach number of
one. From Equation 2, c* is primarily a function of the combustion properties, and from
equation 3, CT is primarily a function of nozzle geometry. To demonstrate the influence
of the fuel-oxidizer composition on c*, Table 2 shows the values for three fuels with
liquid oxygen.
Table 2. Influence on c*7
Oxidizer LO2 LO2 LO2 Fuel LH2 RP1 CH4 O/F 4.83 2.77 3.45 T01 (K) 3250 3700 3560 Avg bulk density (kg/m3) 320 1030 830 c* (m/s) 2386 1838 1783
28
In Table 2 LH2 is liquid hydrogen, RP1 is a hydrocarbon fuel and CH4 is liquid
methane. The values for Table 2 come from Hill & Peterson.7 In each case shown in
Table 2, the O/F ratio was chosen to maximize the Isp and the chamber pressure was 6.89
MPa. As apparent from the data, the molecular weights of the propellants and the values
of c* depend significantly on the fuel-oxidizer combustion.
With Equations 1 through Equation 3, a simple performance analysis can be
conducted. As stated by Huzle & Huang,8 the calculation of thrust-chamber performance
is based on the theoretical propellant combustion data and the application of certain
correction factors. The theoretical propellant combustion data comes from thermo-
chemical computations equating the heat of reaction of the propellant combination to the
rise in enthalpy of the combustion gases. The desired and actual performance of the
chamber calculated from these equations drives the design of the chamber presented next.
2.2.2. Dimension of the Thrust Chamber
The geometry of the thrust chamber is based on the pressure, propellant type,
propellant mass flow rate, and the oxidizer-to-fuel ratio (O/F) derived from the
performance analysis. There are several different approaches to defining the geometry of
the combustion chamber. Humble et.al.6 begins by finding the throat area. The mass
flow at the throat must be chocked for proper operation, therefore the area can be
determined. Through conservation of mass, the throat area (At) can be found from
Equation 4:
ct p
cmA *= (4)
29
To insure long residence times for the mixing and chemical reactions in the thrust
chamber, the Mach number must be nearly zero. The low Mach number implies that the
thrust chamber pressure is nearly the stagnation pressure. Therefore, the chamber area
(Ac) can be found as a multiple of the At, as determined by the thermo-chemistry of the
propellants.6 The result is Equation 5:
( )121
2
211
121 −
+
⎥⎦
⎤⎢⎣
⎡⎟⎠⎞
⎜⎝⎛ −+⎟⎟
⎠
⎞⎜⎜⎝
⎛+
=γ
γ
γγ
MMA
A
t
c (5)
For Equation 5 the Mach number is generally in the range of 0.1 to 0.6. Humble
et.al.6 also derives the chamber volume (Vc) from a ratio of the At. The chamber must be
large enough to allow complete combustion. However, the larger the chamber the greater
the weight. Equation 6 uses a characteristic length (L*) to aid in the sizing of the
chamber:
t
c
AV
L =* (6)
In Equation 6, L* is the thrust chamber’s characteristic length. Historical data
and gas dynamics are the basis for the sizing of L*. Small values of L* imply a small
engine. The goal of the rocket designer is to minimize the size and mass of the engine.
Therefore, L* must be made as small as possible while maintaining adequate combustion
efficiencies.
2.2.3. Injectors
The job of the injector is to atomize, mix, and ignite the fuel and oxidizer.
Ultimately, the injector promotes the complete combustion of the liquid propellants.
Additionally, the pressure loss a cross the injector is important since it takes a way from
30
the pressure obtained in the chamber. To better understand the mechanisms involved in
the injectors, it useful to see what work has been done in this area. Rahman & Santoro9
reported relevant information on predicting spray drop-sizes from liquid oxygen and
gaseous hydrogen. Figure 4 shows a schematic of a coaxial injector.
Figure 4. Shear Coaxial Gas/Liquid Injector
Atomization is of interest to the propulsion community from the standpoint of
propellant injector design for liquid rockets engine combustion chambers. Atomization
from a shear coaxial jet refers to the breakup of the core liquid jet by shear forces due to a
co-flowing, high velocity, annular gas jet surrounding the core liquid stream.9 Data
presented in this paper did not contain any experiments that simulate the unique
propellant properties of liquid oxygen and gaseous hydrogen; such as gaseous hydrogen
injection density and liquid oxygen surface tension and viscosity. Simulation of these
properties is crucial to obtaining reliable atomization results that are directly relevant.
Lightfoot, Danczyk & Talley10 suggested three causes of atomization from wall-
bounded films: liquid turbulence, stripping of waves, and stripping/tearing resulting from
gas-phase vortices. Turbulent eddies within the liquid can interact with the interface
causing it to become roughened and eventually forming ligaments. These ligaments may
then break down into droplets. Results cite several important non-dimensional
parameters including ratios of film height to hydraulic diameter, mean velocity to RMS
31
velocity fluctuations as well as a liquid Weber number based on hydraulic diameter and
mean surface velocity.10 Coherent gas-phase vortices may form as a result of injector
geometry features such as the gap or the lip. If the liquid’s energy is sufficiently larger
than the gas, then the gas-phase vortex will be displaced; otherwise, the vortex will alter
the path of the liquid. In the latter case, droplets may be formed when the vortex distorts
and tears liquid away from the film or when aerodynamic forces arising from the new
shape of the interface strip liquid away.10 Low energy ratio simulations show waves
form on the interface downstream of the lip. These waves are uniform and grow until
they reach a size where the gas flow can strip mass from their crests. The uniform nature
and growth of these waves implies hydrodynamic instabilities cause the droplet
formation. The lip, the spacer between the oxidizer and fuel stream, seems to have very
little effect on the general atomization behavior; a stronger impact is expected if the
gradient of taper is small enough that the flow remains attached to the injector. Current
understanding suggest the relative momentum difference between the liquid and gas
particularly in the axial direction plays a large role in the film’s behavior.10 At high
kinetic energy ratios, gas-phase structures control the film’s behavior and atomization.
At low kinetic energy ratios, waves form on the surface of the film and are responsible
for atomization. Experiment showed atomization due to turbulence occurred at relatively
large gas velocities; suggesting the gas turbulence, not liquid turbulence, was
responsible.10
Besides co-axial injectors, another type of injector that displays beneficial
characteristics is the Pintle injector.11 Figure 5 shows a schematic of a Pintle injector.
32
Figure 5. Pintle Injector
Dressler and Bauer found a Pintle injector design could deliver high combustion
efficiency and enables implementing some unique operating features, such as deep
throttling and injector face shutoff.11 Design simplicity makes it ideal for low cost
engines. Significantly lower development and qualification costs result because injectors
can be easily adjusted and optimized by changing only two simple parts. Additionally,
there has never been an instance of combustion instability in a Pintle engine during any
ground or flight operations.11 Either fuel or oxidizer can be centrally metered in the
Pintle design. The Pintle injector’s flow field induces recirculation regions.
Woodward et. al.12 investigated combustion performance of coaxial and Pintle
injectors using liquid oxygen and ethanol. Though not the exact same as the oxygen and
hydrogen used for the DEAN, the results should show similar trends. The findings shows
marginal liquid oxygen quality was seen to have a significant influence on combustion
efficiency for both injector types.12 Pintle injector designed specifically for orbital
maneuvering system projects operated both in a stable manner and with high
performance. Additionally, changes in O/F ratio resulted in changes in c* efficiency, a
point to consider when optimizing.
33
As shown in this section, many factors affect the design of injectors, however the
work done in this document will be limited to the effect the injector will have on the
combustion efficiency and pressure drop across the injector. Still this information will
be useful in determining the detailed design of the injectors later.
2.3. Nozzle
The key demands on future launch systems are the reduction of Earth-to-Orbit
launch costs in conjunction with an increase in the reliability and operational efficiency.
Launch systems operate in a constantly changing ambient environment. A nozzle that
can maintain high performance over a wide variety of ambient conditions could
dramatically improve efficiency and decrease cost. . The DEAN utilizes one such
concept known as the aerospike or plug nozzle. A booster-stage would experience the
greatest change in ambient pressure and could benefit most form the aerospike design.
However, an upper-stage can benefit as well since it can encounters a change in ambient
pressure depending on where it ignites. Figure 6 show an aerospike.
Figure 6. Aerospike
34
The aerospike nozzle is considered to have globally better performance when
compared to a conventional bell nozzle, since the expansion of the jet is not bounded by a
wall and therefore can adjust itself to the environment by changing the outer jet
boundary.
2.3.1. Aerospike
Hagemann et.al.13 examined several nozzle concepts that could result in superior
performance over conventional nozzles. The advantage of aerospike nozzles is it
demonstrate altitude adaptation up to their geometrical area ratio. This results in
improved performance for the entire flight envelope over the conventional bell nozzle.
Additionally, for an aerospike nozzle, at lower pressure ratios an open-wake flow
established at a pressure level practically equal to the ambient pressure. At a specific
pressure ratio (PR), close to the design PR of a full-length nozzle, the base flow suddenly
changes its character and turns over to a closed form.13 A constant base pressure no
longer influenced by ambient pressure characterizes this flow. Shorter plug nozzles with
high truncations trigger an earlier change in wake flow. At transition, the pressure within
the wake approaches a value below the ambient pressure and the full base area induces a
negative thrust.13 Beyond the transition point, pressure within the closed wake remains
constant, as ambient pressure decreases. The base pressure is then higher than the
ambient pressure, resulting in a positive thrust contribution.13
A more detailed analysis of the aerospike concept was done by Hagemann,
Immich & Terhardt14 using the numerical methods of Euler and Navier-Stokes. The
results showed the altitude compensation capabilities of the aerospike were indisputable,
but they lose this capability for pressure ratios above the design point.14 Additional
35
performance losses were induced due to non-isentropic effects like shock waves. Figure
7 shows the flow over an aerospike nozzle. In Figure 7, the top picture depicts the flow
field at PR below design (lower altitude), the middle picture is at the designed PR, and
the lower picture is at a PR above the design point (higher altitude).
Figure 7. Flow Phenomena of Plug Nozzle
As Figure 7 illustrates, for pressure ratios lower than the design PR of a plug
nozzle with a well-contoured body, the flow primarily expands along the plug body to the
ambient pressure. Thus, only the first part of the nozzle contour acts as an expansion
contour down to the point where the first right running characteristic that feels the
36
ambient pressure meets the contour.14 At the design pressure ratio, the characteristic with
the design Mach number should be a straight line emanating to the tip of the central plug
body, and the shear layer is parallel to the centerline. At pressure ratios above the design
pressure ratio, the wall pressure distribution remains constant, and the plug nozzle
behaves as a conventional nozzle, the loss of its capability of further altitude adaptation
included.14
Truncation of the nozzle makes the aerospike concept more feasible, but results in
further performance losses. Figure 8 shows the flow over a truncated aerospike. In
Figure 8 the top drawing depicts the flow field at PR below design, the middle drawing is
at the designed PR, and the lower drawing is at a PR above the design point.
Figure 8. Flow Phenomena of Truncated Plug Nozzle
37
As seen in Figure 8, at lower pressure ratios an open wake flow establishes, with a
pressure level practically equal to the ambient pressure. At a certain pressure ratio close
to the design pressure ratio of the full-length plug nozzle, the base flow suddenly changes
its character and turns over to the closed form, characterized by a constant base pressure
that is no longer influenced by the ambient pressure.14 At the transition point, the
pressure within the wake approaches a value that is below ambient pressure, and the full
base area induces a negative thrust.14 Beyond the pressure ratio at transition, the pressure
within the closed wake remains constant. A further decrease of the ambient pressure is
resulting therefore in a positive thrust contribution of the total base area.14
Ambient flow also slightly degrades performance. All these losses add to several
percent, making plug perform worse at high altitude when designed for a lower altitude
when compared to similar bell nozzles.14 To get the most benefit, the design pressure
ratio should be chosen at as high an altitude (pressure ratio) as possible.
Concerning the truncation, Ito, Fujii & Hayashi15 sate the nozzle performance was
believed not to be strongly affected by cutting off the nozzle because base pressure
compensated for the loss of the thrust. For a full-length nozzle, flow follows the nozzle
wall and smoothly moves to the wake region. For a truncated nozzle, flow separates at
the trailing edge and expands. The resultant shear layer induces a trailing shock when it
converges and turns the flow parallel to the axis. When truncated, the ramp area
decreases due to the shorter length. Therefore, thrust from the ramp pressure reduces.
The thrust generated by base pressure increases due to the increased area. Therefore, the
total nozzle thrust becomes almost the same for any nozzle truncation.15 Quantitatively,
the thrust coefficients of the plug nozzles have the same trend as the ideal thrust
38
coefficient indicating the plug nozzle operated at near peak thrust efficiency over a wide
pressure ratio range. As the altitude increases, ambient pressure decreases, therefore the
base pressure thrust increases. Ito et.al. have found a contoured plug nozzle had higher
performance compared to a conical plug nozzle over all pressure ratios. As the pressure
ratio increases, the thrust difference between the contoured and conical nozzle increases
almost linearly. The advantage of a contoured nozzle becomes remarkable as the altitude
increases. External flow does not influence the pressure distribution on the nozzle
surface at high-pressure ratios.15
In further work done by Ito & Fujii16, they found that in the low-pressure ratio
regions, the stagnation point was located at a distance from the plug base. When the
pressure ratio increases, the stagnation point suddenly moves toward the plug base. The
stagnation point does not move with further pressure ratio increases. The pressure ratio
at which an abrupt movement occurs is the same as the pressure ratio where the
characteristic transition occurs. The ambient pressure influences base pressure as long as
the flow initiated from the stagnation point due to the envelope shock-wave impingement
reached the base. They conclude the aerospike nozzle performance is insensitive to
length of nozzle because the base pressure compensates the loss of the thrust force due to
nozzle truncation.16 The characteristics of the flow at the base did not change whether
the external flow was induced or not. The ambient pressure influences the base pressure
when the pressure ratio is low and the base pressure becomes independent from ambient
when the pressure ratio is high. The base produces positive thrust when the base pressure
is independent from ambient pressure.
39
In contrast to Ito & Fujii conclusion on external flow effects, Nasuti & Onofri17
found the numerical analyses they performed indicates the interaction of the exhaust jet
with the external air in truncated plug nozzles may significantly affect the nozzle
behavior in the over-expanded regime. In particular, even in the ideal case important
losses take place yielding a sudden performance drop. The presence of a finite-thickness
or thick shroud substantially changes the flow behavior adding further drag in itself and
yielding an overexpansion at the primary nozzle lip (edge of the outer wall). Because of
this overexpansion, the exhaust jet finds a lower ambient pressure at the lip than in the
still air case, and consequently adapts the flow to a lower-than-ambient pressure. This
was a further cause of drag appearing in part over the plug wall and in part over the plug
base. Moreover, the reduction of the value of pressure ratio for transition from open to
closed wake reduced the overall nozzle performance. This reduction yielded a base drag
in the range between the actual and still air transition values. Concerning the mechanism
of wake transition, in the supersonic case the transition is governed by the internal shock,
rather than by the last wave of the expansion fan at the primary nozzle lip.17 On the
contrary, the analysis of the flow-field in the subsonic flight conditions has shown only
slight changes were expected in comparison to the still air case.
Ito & Fujii18 also performed a test to determine the effects of base bleed. They
found the base pressure increased due to an increase in base area and compensated for the
total thrust loss due to a decrease of the ramp pressure thrust. As a result, the total nozzle
thrust becomes almost the same for any nozzle truncation. When they introduced base
bleed, it expanded toward low pressure and interacted with the main exhaust flow at
some distance from base surface. The base bleed that promoted recirculation at the base
40
region produced the largest pressure thrust.18 Additionally, base bleed producing no
divergent loss leads to the maximum total thrust. At low altitude (low PR) base pressure
linearly decreases as the ambient pressure decreases; showing the external environment
influences the base region. Pressure thrust produced by base in this region is small. At
high altitude (high PR), the base pressure is constant despite a variation in altitude
(decrease in ambient pressure). The base-pressure thrust increased in this region. The
base pressure with and without base bleed acted in a similar manner. Conditions with
bleed had a higher-pressure level than without bleed. The thrust coefficients with and
without bleed had the same trends as an ideal thrust coefficient exhibits indicating the
aerospike operates at nearly peak thrust efficiency over a wider range than a bell nozzle.
The thrust performance with base bleed exceeded the performance without base bleed
over the whole altitude range.18 The base bleed orientation with the greatest performance
was at the outer region of the base, directed parallel to the nozzle axis.
2.4. Cooling Jacket
Cooling of the thrust chamber and nozzle is essential in the design of any rocket
engine. Many different cooling approaches exist and some are outlined below.
2.4.1. Heat Transfer
To describe the environment in the thrust chamber, heat transfer equations are
needed. An excellent source to provide an understanding of heat transfer is Incropera
et.al. Fundamentals of Heat and Mass Transfer. 19 The two major mechanisms in which
the heat is transferred are through conduction and convection. Conduction is the
41
transport of energy in a medium due to a temperature gradient.19 A basic equation to
determine the conductive heat rate is shown in Equation 7:
dxdTkAqx −= (7)
In Equation 7 qx is the heat rate with units of watts (W), k is the thermal
conductivity with units W/(m*K), A is the area, and dT/dx is the derivative of the
temperature. The thermal conductivity is a material constant.
Convection describes the energy transfer between a surface and a fluid moving
over the surface.19 A basic equation to determine the convection heat flux is shown in
Equation 8:
( )∞−= TThq s'' (8)
In Equation 8 q’’ is the heat flux with units of W/m2, h is the convective heat
transfer coefficient with units W/(m2K), and Ts and T∞ are the temperature of the surface
and fluid respectively. Unlike the thermal conductivity, which is a material constant, the
convective heat transfer coefficient depends on numerous fluid properties, surface
conditions and flow conditions.
Some of the constants used in the heat transfer equations have to be
experimentally determined. Von Glahn20 proposed correlation treating nozzle heat
transfer with a cooled approach section. This concept applies to both low and high
Reynolds number regions. It also applies to the entire nozzle, whether convergent or
divergent. The proposed correlation consists of a fully turbulent pipe-flow heat-transfer
equation modified by suitable nozzle geometry parameters.20
42
Schacht, Quentmeyer, & Jones21 conducted a gas-side heat-transfer experiment
over a wide range of chamber pressures with an emphasis on the accurate determination
of hot gas-side heat-transfer rates in the regimes of high heat flux. Equation 9 adequately
correlates the data at specific locations in the nozzle:
2.07.0 Re*Pr** −⋅= dCSt (9)
In Equation 9 St*, Pr*, and Re*d are Stanton, Prandtl, and Reynolds number
respectively and C is a constant. The constant varies with axial location and is less than
0.026 for all locations except the chamber. The constant at the throat was 42% less than
the widely used value of 0.026.
Boldman, Schmidt, & Gallagher22 showed the heat transfer at a given station in
nozzles generally exhibits two distinct depressions from the predicted levels based on a
turbulent pipe flow type of correlation. The larger of these depressions occurs at lower
Reynolds numbers and was assumed the result of laminarization of the initially turbulent
boundary layer. The smaller of these depressions, which occurred at high Reynolds
numbers, was assumed the result of reduced turbulent transport associated with a
turbulent boundary layer in an accelerating flow.
An experimental investigation conducted by Quentmeyer & Roncace23
determined the hot-gas-side heat transfer characteristics for a liquid-hydrogen-cooled,
subscale, plug-nozzle rocket test apparatus. The throat region correlation coefficient (Cg)
for a Nusselt number correlation of the form in Equation 10:
3.08.0 PrRegCNu = (10)
43
The averaged value is 0.023 for the Rigimesh faceplate and 0.026 for the platelet
faceplate.
2.4.2. Cooling Techniques
There are several techniques used to cool the chamber and nozzle. The most
common include regenerative cooling, dump cooling, film cooling, transpiration cooling,
and ablative cooling. Regenerative cooling is most relevant to the DEAN and the
technique further studied.
Preclik et. al.24 showed the development of the wall heat flux is not a quick
process, but spreads out from the injector down to the convergent nozzle entrance. The
calorimeter data also highlighted the flow and burning characteristics of the present, co-
axial injection elements. These elements affect the level of the overall wall heat load
approximately 25%. Two principles for reducing the hot gas side heat transfer rates, i.e.
wall element mixture ratio trimming and gaseous hydrogen wall film cooling, were
investigated in more detail. The tests clearly indicated wall film cooling employing the
current slot design is much more efficient and effective when compared to wall element
mixture-ratio trimming. With typical film cooling mass flow rates, the wall heat loads
could be decreased by 20%. Moreover, under the presence of a coolant film, the
individual flow characteristics of the different injection elements become less important.
Kim et. al.25 conducted a test of a regeneratively cooled chamber with film
cooling. From the experimental results with the film cooling mass flow rate at 10.5% of
the main fuel mass flow rate, maximum heat flux at the nozzle throat was measured to be
30% lower than without film cooling at the nominal operating condition. The numerical
analysis resulted in this case showing a 13% decrease. The film cooling mass flow rate
44
increased as the characteristic velocity decreased. When the mass flow rate of film
cooling was 10.5% of the total fuel mass flow rate, the characteristic velocity is measured
to be 1.2% lower than without film cooling at the design point.
In research by Naragi, Dunn, & Coats,26 the effectiveness of regenerative dual-
circuit cooling designs was studied. The dual-circuit showed reduced wall temperatures at
the throat area for the space shuttle main engine (SSME) and for an RP1-LOX engine. A
lower coolant pressure drop accomplished the reduction in wall temperature and resulted
in lower coolant pumping power requirements. The overall result is a higher performing
engine capable of delivering a greater payload to orbit.
Chiaverini, Sauer, & Munson27 used a vortex combustion cold-wall chamber
(VCCW) to study the heat transfer problem. They found the chamber sidewall heating
rates did not display a significant dependence on chamber pressure, apparently due to the
similar effects of elevated pressure on both thermal radiation (acting to heat the wall) and
convective cooling from the outer vortex.
Immich & Caporicci28 conducted an investigations within the FESTIP technology
program on the high aspect ratio coolant channel and micro fin cooling channel chamber.
These technologies have been selected as promising technologies, which can be expected
to lead to considerable reductions of the combustion chamber wall temperature by
improving the coolant side heat transfer significantly. This could be achieved by an
increased heat transfer surface and by taking particular advantage of an enhanced fin
effect. The concept was to increase the total number of cooling channels and to introduce
microstructures at the coolant-side channel bottom.
45
Schmidt, Popp, & Frohlich29 stated an improvement of coolant side heat transfer
and consequently a decrease of the chamber wall temperature has only very minor effect
on the increase of heat flux to the coolant, because the driving temperature difference
cannot be changed much by reducing the wall temperature. A doubling of the coolant-
side heat transfer, by higher coolant velocity, coolant side surface roughening or fins in
the cooling channel etc., would only increase the heat flux to the coolant by about 6-7%.
Increases of heat flux in the chamber by lower chamber diameter and consequently
higher heat flux is another possibility. Reducing the chamber contraction ratio could
increase the heat flux with a resulting 10-15 K (18-27 R) coolant temperature rise, but
only if the chamber is elongated to keep L* constant, which also increases the engine
length.29 The associated increase in cooling and combustion pressure drop would also
increase the required hydrogen pump discharge pressure. This again would reduce the
overall benefit for the cycle to a marginal improvement. On the other hand, from the
injector side a certain face area is required, also limiting the possibility of employing
such a measure. Roughening of the hot gas side surface of the chamber and nozzle would
also increase the heat flux into the coolant. This has been demonstrated by the aging
effect of chambers, running at their thermal limits.29 The most straightforward design
measure for increased heat transfer is the increased surface roughness of the hot gas side
surface.
2.4.3. Aerospike Cooling
The aerospike adds additional challenges when it comes to cooling the nozzle
since it is surrounded by the exhaust flow, whereas the Bell nozzle is exposed to the
ambient surroundings on one side. In Sorge, Carmicino, & Nocito30 a cooled linear plug
46
nozzle was constructed so hot gas tests could be carried out and performance measured.
They chose a two-dimensional nozzle shape and a copper alloy on all the parts exposed to
hot gas flow. In addition, to promote the heat transfer, ribs were placed on the duct
surfaces exposed to the heat flux near the throat. The results have showed the experiment
was able to ensure a maximum wall temperature on the gas side of 750 K (1350 R), a
maximum wall temperature on the liquid side of 450 K (810 R) and, a water bulk
temperature rise of about 298 K (537 R), with a pressure drop per circuit equal to 31
atm.30 Even though this experiment uses water, the technique can be applied to any
cooling fluid.
Kumakawa et.al.31 conducted a test with nickel plating on the hot-gas side. This
test showed the heat flux was 30% lower than without nickel. A cross-flow condition of
water suppressed the burnout on the ribbed surface more than the parallel flow condition.
Combustion performance and heat transfer characteristics for the truncated conical plug
nozzle were quite similar to those of conventional conical nozzles.
From Wang32 the highest level of heating occurs near the thruster outlet, on the
ramp surface. The next highest level of heating occurs on the plug-base where the
reverse-jet brings in the hot plume gases torch the surface. The relief comes from
aspirating cold inner-base airflow into the plug-base region. The airflow penetrates as far
as the base center. The pumping effect of the engine plume causes this aspiration.32 The
heating on the plug sidewall is caused by the hot engine flow spilling off the side of the
ramp, also known as the plume spillage heating.
Tsutsumi et.al.33 observed the maximum heat-flux region is located where the jets
from adjacent modules interact. The heat flux also increased at the module exit. The
47
downstream part of the module exit is exposed to a lower heat load. The heat-flux
distribution over the linear aerospike nozzle is correlated with the viscous-inviscid
interaction in the near-surface flow field. The heat flux increases remarkably at the region
where the module outflows interact due to the downwash of module outflow to the nozzle
surface. Moreover, since the ambient air is entrained into the vortex flow over the nozzle
surface, a lower heat-flux region appears.
2.4.4. Material
Selection of the materials for the cooling jacket tends to drive many aspects of the
design process. Peer & Minick34 describes an advanced combustion chamber designed
by Pratt & Whitney. The most significant feature is the use of a new copper alloy
coupled with an improved processing technique. Their tubes can withstand repeated
exposure to fabrication temperatures in excess of 1250 K (2260 R) and still retain yield
strength five times greater than other copper alloys used in current rocket thrust
chambers.34 The tubular configuration of the chamber provides up to 40 percent more
actual surface area due to the circular tube crowns and therefore, more heat transfer
capability with lower thermal strain (increased life) than smooth wall hot-side fabricated
channel configurations34. The tubular construction also provides improved pressure drop
characteristics over rectangular channel designs.
Schoenman35 discusses the intensive effort to develop materials that can operate at
higher temperatures over the last decade. A nominal target value of 2500 K (4460 R) has
been established for the new materials of construction. The iridium/rhenium layered wall
and the silicon/carbon composite are the most mature and can be considered for flight
48
applications. Iridium/rhenium materials can operate in temperatures 2477 K (4459 R)
while silicon/carbon can operate in 1866 K (3359 R) temperatures.35
Kumakawa et. al.36 showed carbon-carbon and silicon/carbon composites have
potential for use as lightweight materials exposed to temperatures in excess of 1700 K
(3060 R). The exposed heat fluxes dictate whether active cooling would be required for
these materials. Aerospike nozzles made of composite would be exposed to severe
operating conditions in a hot gas flow, including hot spots caused by shock interactions.
Thus, an actively cooled composite is necessary in this case and considered a prime
candidate for the aerospike-nozzle wall structure
A significant challenge will be choosing materials to accommodate the liquid
oxygen expander cycle of this concept. Gloyer, Knuth, & Crawford37 designed a liquid
oxygen cooled gas generator. They found copper, nickel and some steels can be used
adequately with oxygen. These metals are used since they do not react with oxygen that
would cause catastrophic failure. They also concluded the design appears to be limited to
a single operating point design due to the liquid oxygen boiling sensitivity to factors such
as O/F ratio, mass flow, temperature and pressure. For wider operating ranges, alternative
designs which avoid two-phase flow would be preferred.
2.4.5. Aspect-Ratio
Changing the area of the chamber affects the heat transfer characteristics. This
can be done through the aspect-ratio (AR) of the cooling circuit and/or the addiction of
ribs. The heat energy requirements of the turbo-pumps dictate longer combustion
chambers.38 Size limitations create the need for a different method to increase heat
extraction. Increasing the area exposed to the hot-gas by using combustor ribs fulfilled
49
this requirement. The ribs increased the total area exposed to the hot-gas by 80% in his
research, and thus enhanced the heat energy level imparted to the coolant working fluid38.
An investigation was conducted by Carlile39 to determine the effectiveness of
using high-aspect-ratio cooling passages to improve the life and reduce the coolant
pressure drop in high-pressure rocket thrust chambers. Figure 9 show the difference
between a low and high aspect ration channel (AR).
Figure 9. Channel Aspect Ratio
Figure 9 shows that the low AR channel is generally wider with more surface
area along the wall while the high AR is taller with less surface area. The coolant
pressure drop for the high-aspect-ratio chamber was reduced in increments to one-half the
baseline chamber by reducing the coolant mass flow. The result still showed a reduction
in the hot-gas-side wall temperature. The data indicated the hot-gas-side wall
temperatures for the high-aspect-ratio chamber could have been reduced substantially
further by using aspect ratios greater than 5.0.
Wadel & Meyer40 found by increasing the cooling channel surface area through
increasing aspect ratios, heat from the hot-gas-side wall is more efficiently transferred to
50
the coolant. The increased height and number of the ribs also enhance the heat transfer
from the chamber liner to the coolant (i.e. enhanced fin effect). Therefore, it is possible to
fabricate chambers with sufficiently greater total flow area to reduce pressure drop over a
conventional design, and still gain an increase in the heat transfer capability.
In Neuner et.al.41 paper, laboratory experiments with large scale cooling channel
models were described. The purpose of the experiments was to determine the impact of
high-aspect ratio channels on curvature induced heat transfer enhancing phenomena. As
a general result, secondary flow structures were clearly identified, even in channels with
an aspect ratio of 8.0. These vortex phenomena have been found to enhance the heat
transfer both in convex and concave side heated bends. Nevertheless, they appear only in
and not far downstream of the corresponding curvatures.
The results of a parametric study on cooling channel geometry showed as the
channel geometry changes, the coolant heat transfer coefficient dominates the heat
transfer rate as compared to the area terms.42 In general, a small flow area tends to
increase heat transfer; however, the pressure loss across the coolant channel restricts the
extent the area can be reduced. Higher aspect ratio cooling channels are advantageous in
balancing the pressure loss requirements with the heat transfer demands. The maximum
allowable chamber pressure is limited by the survivable gas wall temperature. With
respect to pressure loss and over all engine mass, a stepped-channel configuration proves
to be superior to an invariant channel while maintaining the thermal performance of the
regenerative cooling jacket. Evidence of this was presented by Schuff et.al. and is shown
in Table 3.
51
Table 3. Stepped Channel vs. Constant Cross-Section Channel42
Constant Cross- Section Stepped Pin (psi) 2299 1838 Pout (psi) 1762 1680 ΔPc (psi) 537 158 ΔTc (R) 766 737 Engine Mass (lbm) 327 274
Table 3 shows a ΔPc and ΔTc decreased of 340% and 4%, respectively, from the
constant cross-section to the stepped channel. Schuff et.al.42 state near the throat the gas
side heat transfer coefficient increases significantly and as a result the wall temperature
also increases. To adequately deal with the high heat transfer rate, the geometry remained
the same as defined for the baseline case in this region. In the reduced combustion gas
heat transfer regions upstream and downstream of the throat we increased the channel
cross-sectional area to decrease the pressure drop and maintain a combustion gas side
wall temperature less than at the throat. Consequently, increasing the channel cross-
sectional area, decreased the coolant heat transfer coefficient hc, therefore the stepped-
channel case presented a tradeoff in performance parameters, temperature rise (ΔTc) and
pressure drop (ΔPc) of the coolant along the entire length of the channels.
52
3. Methodology
This section will outline the methodology implemented to design the thrust
chamber, nozzle, and cooling jacket of the DEAN. To begin the process, the
performance goals outlined in the introduction establish a baseline engine derived from
basic rocket equations.
3.1. General Sizing
The design started with the determination of a solution space for the size of the
engine. The size of the engine begins with the choosing of initial parameters; the fuel
and oxidizer, the oxidizer-to-fuel ratio (O/F), the chamber pressure (Pc), and expansion
ratio (ε). As stated in the introduction, the DEAN will be powered by hydrogen and
oxygen. Table 4 shows the remaining initial values to start the process.
Table 4. Chosen Parameters
O/F ε Pc (MPa/psi) 7 125 12/1740
The values in Table 4 are chose to maximize the Isp of the design. The pressure
and expansion ratio were chosen based on other like designs such as the RL-10. The exit
Mach number (Me) was found using ε and the ratio a specific heat (γ) in Equation 11:
53
221
2
211
121 −
+
⎥⎦
⎤⎢⎣
⎡⎟⎠⎞
⎜⎝⎛ −+⎟⎟
⎠
⎞⎜⎜⎝
⎛+
=γγ
γγ
ε ee
MM
(11)
The value for γ is for a combusted flow and was assumed at 1.2 for the initial
estimates. The exit Mach number along with the chamber pressure and γ were used in
Equation 12 to determine the exit pressure:
γγ
γ −
⎥⎦⎤
⎢⎣⎡ −+=
12
211 ece MPP (12)
Based on the chemical reaction, the characteristic exhaust velocity (c*) can be
found using Equation 13:
221
*
12
*−+
⎟⎟⎠
⎞⎜⎜⎝
⎛+
=γγ
γγ
γη cc RTc (13)
In Equation 13, the chamber temperature and the gas constant come from the
chemical kinetics of the hydrogen and oxygen. The c* efficiency (ηc*) is set to a realistic
value based on flow conditions and empirical evidence. From the results of Equation 11-
13, the theoretical Isp can be found using Equation 14:
( )⎪⎭
⎪⎬
⎫
⎪⎩
⎪⎨
⎧
−+⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
⎟⎟⎠
⎞⎜⎜⎝
⎛−⎟⎟
⎠
⎞⎜⎜⎝
⎛+⎟⎟
⎠
⎞⎜⎜⎝
⎛−
=
−
−+
aecc
e
osp PP
Pgc
PP
gcI
0
111
*11
21
2* εγγ
γλγγ
γγ
(14)
In Equation 14, λ is the nozzle efficiency and is set to a reasonable number based
on other models. The ambient pressure is set to zero for a vacuum providing a maximum
value. With the Isp and the thrust, the mass flow can be calculated from Equation 1. The
geometry of the combustion chamber can then be calculated as outlined is section 2.2.2.
54
3.2. Thrust Chamber
The purpose of the thrust chamber is to combine and combust the fuel and
oxidizer. Several variables are involved in the design of the thrust chamber. Below are
the input variables that cannot be changed (controls), variables that can be changed by the
designer to produced the desired results (process variables), and the results used to
determine the validity of the design (outputs).
3.2.1. Chamber Controls
The constant variables in the design of the chamber are the mass flow rate of the
hydrogen and oxygen. The values for these mass flow rates come from a combination of
the requirements of the overall cycle balance and sizing.
3.2.2. Chamber Process Variables
The changing variables include several key geometric aspects of the chamber.
The primary aspects are the radius and length of the chamber. In the NPSS model, these
variables are the radius and volume of the chamber. Additionally, the contours of the
chamber can change to get desired results. These include the outer chamber wall and the
inner spike
3.2.3. Chamber Outputs
The output variables used to evaluate the results are the pressure and temperature
of the chamber. The pressure and temperature must be within material limits so the
structural integrity of the chamber remains intact. However, with higher values, the
performance of the rocket is generally better.
55
3.3. Nozzle
The purpose of the nozzle is to expand the flow and produce thrust. Below are
the controls variables, process variables, and output results affecting the design of the
nozzle.
3.3.1. Nozzle Controls
The variables held constant in the design of the nozzle are the nozzle outputs.
These include the pressure, temperature, and mass flow.
3.3.2. Nozzle Process Variables
The variables that change are again the geometry of the nozzle. For the nozzle,
this includes the throat area, ε, and the length. Additionally, the operating altitude is set
by varying the ambient pressure input. The throat area is constrained by the area of the
chamber and the limit of the Mach number at the throat equal to one. However, the area
of the throat can be changed to accommodate both the chamber and nozzle but must
maintain the mass flow.
3.3.3. Nozzle Outputs
The overall performance statistics characterizing the success of this effort are the
Isp and the thrust. The desire is to have the Isp and thrust close to, if not more than the
goal values listed in Chapter 1.
3.4. Cooling Jacket
The cooling jacket cools the walls of the thrust chamber and transports the proper
energy to the turbines. Below are the control variables, process variables, and output
results affecting the design of the cooling jacket.
56
3.4.1. Cooling Jacket Controls
The constant set for the cooling jacket come from the requirements of the turbo-
machinery. These include the mass flow rate of the coolant, the inlet temperature, and
the inlet and outlet pressures.
3.4.2. Cooling Jacket Process Variables
The process variables to be examined in the design space include the geometry of
the cooling jacket. In NPSS, these are the area, volume, and number of channels.
Another important parameter includes the surface area of the cooling jacket that transfers
the heat. The geometry of the cooling jacket is dependent on the geometry of the
chamber and nozzle. However, these geometries, such as the cross-sectional area of the
channels, can be modified to meet the requirements of the cooling jacket.
3.4.3. Cooling Jacket Outputs
The analysis will focus on the wall and outlet temperatures of the cooling jacket
in evaluating the performance. The wall temperature must remain low enough to prevent
the wall from melting while the outlet temperature must be high enough to ensure
sufficient power is transferred to the turbines.
3.5. NPSS
After defining the performance parameters of the engine, the data was input into
NPSS to perform a power balance and pressure budget. The power balance and pressure
budget confirms the validity of the DEAN design. Additionally, the NPSS codes has a
high fidelity resulting in more accurate compared with the initial estimates. Additionally,
changes to the design can be made quickly with the effects of the changes apparent.
57
NPSS was developed by NASA Glenn Research Center, in conjunction with the
U.S. aero propulsion industry and the Department of Defense, to grow technologies
capable of supporting detailed aerothermomechanical computer simulations of complete
aircraft engines. NPSS can realistically model the physical interactions taking place
throughout an engine, accelerating the concept-to-production development time and
reducing the need for expensive full-scale tests and experiments.43
A rocket engine was built in NPSS using elements to define the various
components. Figure 10 shows a schematic layout of the NPSS model used to define the
DEAN that is based on the schematic to the DEAN show in Figure 1.
Figure 10. NPSS Model Schematic
As Figure 10 illustrates, each block represents an element and the elements are
linked to one another through link ports. NPSS comes with defined elements for both air
58
breathing and rocket engines. Additionally, new elements can be programmed by the
user. Each element has a set of input/output variables, independent/dependent variables,
and equations. The user defines the input variables, though not all inputs need to be
defined. Additionally, in the model, the composition of the flow, such as oxygen, and the
thermodynamic data pack are defined. The inputs defined by the user, and the inputs of
the composition and thermodynamic data, are used by the equations, with changing of the
independent and dependent variables as defined in the element, to determine the outputs.
The elements are linked through linkPorts. There are link ports for fluid and heat flow.
Each element defines which type, and how many ports the element has.
The full model used for the DEAN is presented in Appendix B. For the
combustion chamber, labeled COMB in Figure 10, was modeled using the RocketComb1
element. This element requires inputs of chamber radius, chamber volume, and the
composition of the oxidizer and fuel inlet, and outlet. Additionally, guesses for the
oxidizer-fuel ratio (OFR), chamber temperature, chamber pressure, and weight flow are
required. The aerospike nozzle, labeled NOZZ in Figure 10, is modeled with the element
RocketNozzle and requires inputs of the throat area, area ratio (ε), and ambient pressure.
The cooling jacket model requires the use of three elements; one for a pressure
drop, one for change in energy, and one for the transfer of thermal energy, because of the
way NPSS works. Figure 11 shows one of the cooling jackets of the NPSS model.
59
Figure 11. NPSS Cooling Jacket
The Valve04 element modeled the pressure drop of the cooling fluid. The
elements DuckH2 in Figure 11 is an example of a Valve04 element in the DEAN model.
The Valve04 element requires an input of the cross-sectional area and friction factor.
Furthermore, a guess at the weight flow is required. The CoolingVolume element was
used to model the change in energy of the flow. The elements CVH1 and CVH2 in Figure
11 are examples of this element in the DEAN model. The CoolingVolume element
requires inputs of the volume, area, and number of channels. In addition, a guess at the
total pressure and total specifies enthalpy. The Wall2 element was used to model the heat
transfer from the chamber to the cooling fluid. The elements Hwall1 and Hwall7 in
Figure 11 are examples of this element in the DEAN model. This element requires and
input of the specific heat (cp), thermal coefficient (k), and the density. Additionally, a
guess of the wall temperature is required.
Other elements used in the DEAN model are the Starter, Pump, Turb02, and Shaft
elements. In Figure 10, OTNK is an example of a Starter element. These elements begin
and end mass flow respectively. These elements require a temperature and pressure as
inputs. The Pump element models a fluid pump and is used for such elements as OP in
Figure 10. This element requires and input of weight flow, gear ratio, efficiency, and
60
pressure rise. To model a turbine the Turb02 element was used. One such element
shown in Figure 10 is HT. This element requires an input of weight flow, efficiency, and
pressure ratio. The Pump and Turb02 element are linked by the Shaft element which
balances the power. In Figure 10, HSHAFT is a Shaft element. This element only
requires and input of the revolutions per minute (rpm).
To allow mass flow from one to another, elements are linked with
UnReactedFluidPorts. These ports do not require any specific imports, however the user
may define a parameter. Heat is passed between elements using ThermalPorts. Inputs
for these ports include a radius of curvature, cross-sectional flow area, and a surface area.
3.6. TDK
The Two-Dimensional Kinetics (TDK) code was used to establish the contour of
the aerospike nozzle. The results of the this code will aid in the inputs for the nozzle in
NPSS and help confirm the results obtained.
The TDK computer program represents the culmination of work done by many
people over the last 35 plus years. A series of contracts funded through NASA and the
ICRPG (now (JANNAF) resulted in TDK becoming the JANNAF Standard Code for
predicting the nozzle performance in liquid propellant rocket engines in 1967.44
Currently, TDK is property of and under development by Software and Engineering
Associates, Inc. (SEA).
The TDK program is composed of seven modules, ODE, SCAP, ODK, TRANS,
MOC, BLM, and MABL. These modules allow for a complete two dimensional non-
61
equilibrium nozzle performance calculation with a boundary layer. Each module requires
one set of inputs and can be used alone, in part, or in combination with the other modules.
3.7. Sensitivity Analysis
A sensitivity analysis on the cooling jacket was performed to focus the design
changes in the cooling jacket parameters. To accomplish this, a simplified NPSS model
was created with a single wall circuit for both hydrogen and oxygen flow. Figure 12
show the schematic of the sensitivity model.
Figure 12. Sensitivity Model
The primary outputs that would be used to judge the merit of the design are the
final temperature of the flow and the temperature of the wall. The final temperature must
be great enough to allow the turbine to produce the required amount of horsepower to
drive the pumps. The wall temperature must be sufficiently low to insure structural
stability of the material. The parameters that were controlled are the mass flow, change
62
in pressure, cross-sectional area, and surface area. Equation 15 was used to compute the
sensitivity of each parameter:
QQTySensitivit
ΔΔ
= (15)
In Equation 15, Q represents the parameter that was changed and T is the
temperature. Table 5 and Table 6 show the initial values and the results of the sensitivity
analysis.
Table 5. Hydrogen Sensitivity Analysis
Hydrogen Original New Tout initial
Tout new
Twall initial
Twall new
Tout sensitivity
Twall sensitivity
mdot 10.0 9.06 221.6 223.7 729.7 766.2 -22.1 -382.5ΔP 500 700 221.6 221.2 729.7 730.2 -1.18 1.25Cross Area 1 1.5 221.6 220.5 729.7 916.9 -2.28 374.4Surface Area 50 55 221.6 223.9 729.7 729.1 22.4 -6.51
Table 6. Oxygen Sensitivity Analysis
Oxygen Original New Tout initial
Tout new
Twall initial
Twall new
Tout sensitivity
TWall sensitivity
mdot 10.0 8.44 281.1 285.1 1656 2066.1 -25.9 -2629.4ΔP 500 700 281.1 281.2 1656 1665.5 0.313 23.75Cross Area 1 1.5 281.1 280.4 1656 1780.6 -1.38 249.14Surface Area 50 55 281.1 282.8 1656 1855.6 17.7 1995.6
The initial values in Table 5 and Table 6 were chosen as realistic round numbers
for ease of analysis. The highlighted boxes are those that affect the temperature of the
fluid and wall the most. From Table 5, a change in mass flow results in a proportionally
63
large change, in the opposite direction, in both the flow and wall temperature.
Additionally, a change in cross-sectional area results in a substantial change in wall
temperature in the same direction. A change in surface area results in a considerable
change in the flow temperature in the same direction.
From Table 6, a change in oxygen mass flow has a similar effect on the flow and
wall temperature as the hydrogen mass flow. Furthermore, a change in the surface area
has a significant change in both flow and wall temperature in the same direction.
Plainly, the mass flow, cross-sectional and surface area have the largest effect on
the final temperature of the flow and the wall temperature.
3.8. Beginning Frame Work
The inputs to the NPSS model require many geometric inputs. Therefore, a
general frame for the chamber was developed to begin the process of defining the
geometry inputs for NPSS. Figure 13 shows the chamber framework.
Figure 13. Initial Contour
The contour illustrated in Figure 13 helped to visualize the design of the chamber.
With the framework of the chamber and nozzle; the geometry, such as the radius and
length, could be adjusted based on the needs of the design. The changes would then be
used to determine inputs into NPSS such as the surface area for a ThermalPort.
64
3.8.1. Cooling Channel Design
Another aspect affected by the contour of the chamber is the design of the cooling
channels. The number and size of the cooling channels is constrained by the radius of the
chamber at each point. Initially both milled channel and tubular channels were
considered but the milled channels proved to be superior in performance and
manufacturability. Figure 14 shows the cross-section used to define the cooling channel
geometry.
Figure 14. Cooling Channel Cross-Section
The channel depicted in Figure 14 represents half of the actual channel where it is
mirrored about the centerline. The shaded region represents the solid material while the
clear area is where the fluid flows. The parameter ’a’ represents the half spacing between
channels, ‘w’ is the half width, ‘t’ is the thickness between the chamber wall and the
channel, ‘h’ is the height of the channel. To determine the number of channels that will
fit, the circumference of the chamber is divided by twice the sum of ‘a’ and ‘w’, since
one channel is represented by two times their lengths.
65
Changing the thickness directly influences the hot and cold side wall-temperature.
The ratio between ‘a’ and ‘w’ determines the total surface area of the channels thereby
influencing the wall temperature and the temperature of the fluid as indicated in the
sensitivity analysis. Changing the height of the chamber affects the cross-sectional area
of the channel thus affecting the wall and fluid temperature as well.
3.8.2. Material Choice
The methodology for choosing the materials was based on the properties
presented here. The material of the chamber must deal with extreme thermal and
pressure forces. To handle the heat, materials with high thermal conductivity and
elevated melting points were considered. These include metals like copper and nickel as
well as composites like silicon carbide. Table 7 shows a list of materials and their
important properties.
Table 7. Material Thermal Properties19
Material
Melting Point (R)
50% of Melting
(R)
Thermal Conductivity (Btu/in*R*s)
Copper 2444.4 1222.2 0.005377 Beryllium 2790 1395 0.002682 Chromium 3812.4 1906.2 0.001257 Cobalt 3184.2 1592.1 0.00133 Iridium 4896 2448 0.001971 Molybdenum 5209.2 2604.6 0.001851 Nickel 3110.4 1555.2 0.001216 Rhodium 4024.8 2012.4 0.002012 Silicon 3033 1516.5 0.001985 Silver 2223 1111.5 0.005753 Tungsten 6588 3294 0.002333 Niobium 4933 2045 0.00072 Silicon Carbide 5580 2790 0.006571 Carbon/Carbon 4091.4 5045.7 0.02615
66
The 50% melting point temperature in Table 7 will be the values the wall
temperature is compared to in order to determine if the material is viable given that the
strength of the material should remain unchanged up to that point.
Surrounding the cooling jacket will be a structural jacket to withstand the
pressure. Materials for this task will need high tensile strength to handle the pressure and
low density to minimize the weight. Table 8 displays materials considered for structural
support.
Table 8. Structural Materials
Material Ftu (ksi) Density (slugs/ft3)
Al 7075-T6 80 5.4Copper 33 17High-strength Steel 180 15.2Nickel 110 17Titanium 170 8.7Tungsten 600 37Carbon Fiber 30.5 2.72
The values in Table 8 are average values and for materials like carbon fiber, are
highly dependent on the specific usages and manufacture. Another consideration for the
material choice is how it reacts with oxygen. According to a NASA report,45 materials
used in an oxygen environment should have a low heat of combustion and a
comparatively low burn factor. Equation 16 shows how burn factor can be defined:
α
0fH
BFΔ
= (16)
In Equation 16, ΔH0f is the heat of oxidation and α is the thermal diffusivity.
Table 9 shows to compare material compatibility with oxygen.
67
Table 9. Oxygen Compatibility
Material
Heat of Combustion45 (cal/g)
Heat of Formation46 (kcal/gmole)
Thermal Diffusivity19 (m2/s)
Burn Factor
Copper 585 41.8 117 0.357 Aluminum 7425 400.4 97.1 4.12 Cobalt 970 57.1 26.6 2.15 Molybdenum 1458 182.65 53.7 3.40 Nickel 980 57.3 23 2.49 Tungsten 1093 140.94 68.3 2.06 Titanium 4710 225.5 9.32 24.19
Aluminum and Titanium are metals known to be highly reactive with oxygen.
Copper and Nickel are generally unreactive with oxygen. The values in Table 9 for these
materials are used as baselines to evaluate other materials. For example, Tungsten’s burn
factor is close to copper, therefore it would be considered compatible with oxygen. On
the other hand, Molybdenum has a burn factor closer to aluminum and consequently
would be considered to be reactive with oxygen. On the composite side, Schoenman45
states that silicon carbine is compatible with use in oxygen environments.
68
4. Analysis and Results
4.1. Chapter Overview
This Chapter presents the results obtained from NPSS and TDK for the DEAN
described in the methodology of Chapter three. The NPSS model of the DEAN consists
of five major components; the oxygen turbo machinery, the oxygen cooling jacket, the
hydrogen turbo machinery, the hydrogen cooling jacket, and the chamber/nozzle. The
turbo machinery is analyzed in related documents by Capt Michael Arguello47 for the
hydrogen side and Capt William Strain48 for the oxygen side. Presented below are the
results pertaining to the cooling jackets and the chamber/nozzle. The code that produced
these results is presented in Appendix C.
4.2. Results of Simulation Scenarios
The chamber and nozzle contour was defined to determine the geometric inputs
for NPSS and visualize the deign of the DEAN. Figure 15 shows the final contour of the
chamber/nozzle used in the NPSS model.
Figure 15. Chamber/Nozzle Contour (all dimension in inches)
69
Figure 15 shows the chamber/nozzle split into several segments. Splitting the
contour into segments allowed for a profile of the wall and fluid temperature along the
length of the contour. The chamber is split into five equal length sections with an
additional section at the throat since that point usually experiences the most extreme
temperatures. The nozzle is split into three sections with the two equal length sections
being cooled and the last section not cooled. The mid-points shown in Figure 15
represent the points where the geometry for the elements are defined for each section.
4.2.1. Hydrogen Cooling Jacket Results
The hydrogen cooling-jacket performs the task of cooling the inner chamber wall
and part of the nozzle while providing energy to the hydrogen turbine to power the
pumps. The hydrogen flow is split 50/50 after the first pump as shown in Figure 10. The
result is a mass flow of 7.55 lbm/s (3.42 kg/s) for cooling. The cooling jacket consists of
the elements between the second pump and turbine. The DuckH# elements are Valve04
elements and are responsible for the pressure drop in the jacket. The pressure drops from
4050 psi (27.9 MPa) at the exit of the second pump to 3625 psi (25 MPa) at the inlet of
the turbine. The friction factor (K) of the elements is set to allow for a constant pressure
drop through the jacket.
The CoolingVolume elements attracted to the Wall2 elements allow for the heat
transfer from the chamber to the fluid. These elements consist of CVH# and Hwall#
elements. There are a total of 314 channels with the dimensions of the channels changing
for each element as revealed in Table 10.
70
Table 10. Hydrogen Channel Dimensions
Element r (in) a (in) w (in) h (in) A (in2) AR CVH1 2.5 0.01 0.015 0.02 0.19 1.33 CVH2 4 0.01 0.03 0.02 0.38 0.67 CVH3 4.5 0.01 0.035 0.02 0.44 0.57 CVH4 4 0.01 0.03 0.02 0.38 0.67 CVH5 2.75 0.01 0.0175 0.03 0.33 1.71 CVH6 2.25 0.01 0.0125 0.04 0.31 3.2 CVH7 2 0.01 0.01 0.05 0.31 5 CVH8 2 0.01 0.01 0.05 0.31 5
The variables in Table 10 correspond to Figure 14, r is the radius and A is the
total cross-sectional area of the cooling flow. The change in dimension in Table 10
represents the change in priority of the channel along the contour of the chamber. At the
throat, CVH3, the AR is lowest to allow for maximum heat transfer and minimum wall
temperature. At the end of the flow, CVH8, the concern is preventing the flow from
going supersonic. A shock wave in the channel would greatly increase the pressure loss
and may also cause damage to the structure. Table 11 shows the Mach number for each
element.
Table 11. Hydrogen Mach Numbers
Element P (psi) T ( R ) V (ft/s) a (ft/s) Mach # CVH1 4002 145.3 1630 4556.4 0.35 CVH2 3954 270.6 1336 4338.5 0.30 CVH3 3907 276.6 1185 4333 0.27 CVH4 3860 415.7 1972 4689.1 0.42 CVH5 3813 477.2 2584 4870.7 0.53 CVH6 3766 524.9 3029 5012.1 0.60 CVH7 3719 567.4 3278 5137.1 0.63 CVH8 3672 609.7 3532 5260.7 0.67
In Table 11, the pressure, temperature and velocity are the exit properties of each
element. The values for speed of sound (a) come from the National Institute of Standards
71
and Technology (NIST)49 database for hydrogen for the specific pressure and
temperature. The greatest Mach number is 0.67 at the end of the cooling jacket and is not
high enough to cause any issues associated with shock waves. The Mach number is not a
issue for hydrogen till around 0.9, therefore there is some leeway were this is concerned.
The heat flow and heating coefficients used to calculate the wall temperature are
outputs from the NPSS model. However, the wall temperature NPSS calculates is only a
balance of heat in and out of the wall and does not take into account material properties.
To calculate the wall temperature for different materials another calculation is required.
The calculation balances the three heat equations shown in Equation 17 originating from
Equations 7 and Equation 8:
)()( CWCCCWCWH
HWHHHH TTAht
TTkATTAhq −=⎟
⎠⎞
⎜⎝⎛ −
=−= (17)
In Equation 17, q, hH, hC, AH, AC, TH, and TC come from the NPSS model. The
value for k is dependent of the material chosen, and t is set to 0.02 in (0.508 mm) for the
design of the cooling jacket. The results for various materials are presented in Appendix
E. Figure 16 shows the graph of the wall temperature for copper along the axial
coordinate.
72
Figure 16. Hydrogen Wall Temperatures
In Figure 16, the injector faceplate is at zero and the beginning of the cooling
circuit in the nozzle is at 31 in (0.787 m) (the hydrogen travels from right to left). The
throat is at 24 in (0.61 m). As expected the hottest wall temperature is at the throat at
1224 R (680 K). The 50% melting point for copper is 1222 R (678 K). The calculated
temperature of the wall is within a percent of the 50% melting point of copper, hence in
the current configuration the wall temperature maintains a reasonable level during normal
operation.
The final requirement for the cooling jacket to meet is the temperature of the flow
entering the turbine. Table 12 shows the outlet temperature of each element.
73
Table 12. Hydrogen Flow Temperature
Element T (R) CVH1 145.3CVH2 270.6CVH3 276.6CVH4 415.7CVH5 477.2CVH6 524.9CVH7 567.4CVH8 609.7
Table 12 shows the temperature into the cooling jacket is 145.3 R (80.7 K) at the
subsequent rise in temperature to The result is a temperature raise of 464.3 R (257.9 K).
The temperature into the turbine is 609.7 R (338.7 K) that allowed the turbine to produce
3573 hp, adequate to power the hydrogen pumps.
4.2.2. Oxygen Cooling Jacket Results
The oxygen cooling-jacket is responsible for cooling the outer chamber wall and
providing energy to the oxygen turbine to power the pump. Unlike the hydrogen, 100%
of the oxygen flows through the jacket as illustrated in Figure 10. The result is a mass
flow of 106 lbm/s (48.08 kg/s) for cooling. The cooling jacket consists of the elements
between the pump and turbine. The DuckO# elements perform the same task of pressure
drop like the DuckH# elements. The pressure drops from 4635 psi (31.9 MPa) at the exit
of the pump to 3666 psi (25.2 MPa) at the inlet of the turbine.
The elements of CVO# and Owall# elements perform the same as CVH# and
Hwall# elements. The oxygen cooling-jacket has 785 channels with the dimensions of
the channels changing for each element as revealed in Table 13.
74
Table 13. Oxygen Channel Dimensions
Element r (in) a (in) w (in) h (in) A (in2) AR CVO1 5 0.01 0.01 0.03 0.47 3 CVO2 5.5 0.01 0.012 0.04 0.75 3.33 CVO3 6 0.01 0.014 0.04 0.88 2.86 CVO4 6 0.01 0.014 0.04 0.88 2.86 CVO5 6 0.01 0.014 0.04 0.88 2.86 CVO6 6 0.01 0.014 0.05 1.1 3.57
The variables in Table 13 are the same as Table 10. The change in dimension in
Table 13 are due to the same factors that influence the change in AR seen in Table 10.
Table 14 shows the Mach number for each element.
Table 14. Oxygen Mach Numbers
Element P (psi) T ( R ) V (ft/s) a (ft/s) Mach # CVO1 4497 179.3 448.2 3186.3 0.14 CVO2 4359 303.9 379.8 1980.1 0.19 CVO3 4221 379.6 428.1 1497.1 0.28 CVO4 4084 452.2 576.2 1326.1 0.43 CVO5 3947 531.4 758.3 1321.6 0.57 CVO6 3809 616.8 764.3 1354.6 0.56
In Table 14, the pressure, temperature and velocity are the exit properties of each
element. The values for speed of sound are for oxygen at the specified pressure and
temperature. The greatest Mach number is 0.57 at CVO5 and is not high enough to cause
any issues as stated earlier with the hydrogen side. The Mach number becomes an issue
at around 0.6 for oxygen. Therefore, this design is at the high end of what is expectable.
The wall temperature for the oxygen side is computed using Equation 17 similar
to the hydrogen side. The results for various materials are presented in Appendix E.
Figure 17 shows the graph of the wall temperature for silicon carbide along the axial
coordinate.
75
Figure 17. Oxygen Wall Temperatures
In Figure 17 the injector faceplate is at 0 inches and the beginning of the cooling
circuit in the throat is at 24 inches (0.61 m) (the oxygen travels from right to left).
Silicon carbide is an expensive and exotic material, however, lower cost alternatives we
unable to meet the performance requirements. Additionally, silicon carbide is know to
works well with oxygen, that was a consideration outlined in section 3.8.2. Silicon
carbide has been successfully tested with an actively cooled configuration.50 Therefore, it
is manufacturable although it may be expensive. Another material that might work is
tungsten. According to the burn factor presented in Chapter three, tungsten should be
compatible with oxygen but the manufacture may present problems.
As expected the hottest wall temperature is at the throat at 1625 R (902 K). The
50% melting point for silicon carbide is 2790 R (1550 K). The calculated temperature of
the wall is 34 % lower than the 50% melting point of silicon carbide; hence, in the current
76
configuration the wall temperature maintains a reasonable level during normal operation
and allows room for growth.
The final requirement for the cooling jacket to meet is the temperature of the flow
entering the turbine. Table 15 shows the outlet temperature of each element.
Table 15. Oxygen Flow Temperature
Element T (R) CVO1 179.3CVO2 303.9CVO3 379.6CVO4 452.2CVO5 531.4CVO6 616.8
Table 15 reveals the temperature into the cooling jacket is 179.3 R (99.6 K) and
increases along the jacket to a final temperature of 616.8 R (342.6 K). The result is a
temperature rise of 437.4 R (243 K) in the jacket. The temperature out of the cooling
jacket allowed the turbine to produce 2587 hp, sufficient to power the oxygen pump.
4.2.3. Chamber/Nozzle Results
The hydrogen and oxygen flow into the chamber and combust, and then expand
through the nozzle. The elements of interest are the COMB and NOZZ elements as
depicted in Figure 10. The COMB element has an input of C_O2_H2 as the composition
to designate that the reaction is of hydrogen and oxygen and the radius of the chamber is
6 inches. The NOZZ element has inputs of 15.9 in2 for the throat area, 125 for the
expansion ratio, and 0.1 psi for the ambient pressure. Both elements have thermal ports
that are connected to Wall2 elements for heat transfer. The inputs for radius of curvature,
77
flow area, and surface area (areaHx in the code) come from Figure 15. The results for
these elements are presented in Table 16.
Table 16. Chamber Performance
O/F 7.03 Pc 1739.1 psi 12 MPa Tc 6586.3 R 3659 K Fg 57231.9 lbf 254 kN Total mass flow 121.1 lbm/s 54.9 kg/s Hydrogen mass flow 15.1 lbm/s 6.8 kg/s Oxygen mass flow 106 lbm/s 48.1 kg/s Isp 472.7 sec
Table 16 show this model meets the thrust and Isp goals set forth in Chapter 1.
Specifically, the thrust goal is 50,000 lbf (222.4 kN), while the result from the DEAN
NPSS model is 57,231 lbf (254 kN). This is a 14% increase in thrust over the goal. The
Isp goal is 464 sec whereas the DEAN Isp from the NPSS model is 472.7 sec. This is a
1.9% increase in Isp over the goal. The increase of the thrust and Isp over the goal values
shows the model performs better than expected. Additionally, the values for the chamber
are close to the initial parameters shown in Table 4. The O/F is 7.03 in the model while
the initial value was seven. The chamber pressure of 1739 psi (12 MPa) is within one psi
of the initial selection. This demonstrates the initial assumptions were sound and the
NPSS chamber model performs close to what was projected.
4.2.4. TDK Nozzle
TDK 04 determined the contour of the nozzle. The code developed for the DEAN
is given in Appendix C. In developing the code, the goal was to match many of the
chamber properties, such as pressure and temperature, to the NPSS model. TDK then
generated a nozzle based on these properties and non-dimensionalized by the throat
78
radius. The contour is then incorporated into the design. The code developed results in
the following parameters in Table 17.
Table 17. TDK Properties
Pc 1739 psi 12 Mpa Tc 6586 R 3659 K O/F 7 Mdot 121.1 lbm/s 54.9 kg/s Fg 66167 lbf 294 kN Isp 546 s
Table 17 illustrates the chamber properties for NPSS and TDK are the same.
However, the Isp and thrust calculated by TDK is much larger than the NPSS model.
Some difference is due to the differences in methods of TDK and NPSS. Nevertheless,
the TDK Isp is still above the theoretical limit of 500 sec for an O2/H2 engine. This is
explained by the fact that the presented Isp is that of the entire control volume TDK used
to make calculations and therefore is inflated. Still, the definition of the control volume
in unclear in the output so real Isp remains unknown. The nozzle contour was still used
for NPSS since the mass flows matched. Figure 18 shows the non-dimensional contour
with a temperature profile.
Figure 18. Non-dimensional Nozzle Contour with Temperature Profile
79
Figure 18 reveals the length of the nozzle is about three times the radius at the
throat. Additionally, Figure 18 shows the change in temperature along the nozzle contour
and is the reason for part of the nozzle being actively cooled due to the decrease in the
temperature of the flow.
Truncation is ending the nozzle before the flow is fully expanded. Figure 19
show a truncated nozzle.
00.5
11.5
22.5
33.5
44.5
5
0 2 4 6 8 10 12 14
L (in)
R (in
)
Figure 19. Truncated Nozzle
In Figure 19, the red line denotes a length of 75% of a fully contoured nozzle, the
green line 50% and the black like 25%. To determine the affect of truncation, a model
where the ratio of length to throat radius varies was created. Table 18 shows the relation
of nozzle length to Isp and mass.
Table 18. Truncation of Nozzle
L Isp (s) ΔIsp (s) Mass (lbm) ΔMass 100% 546 1.05 75% 548 2 0.96 8.5% 50% 545 -1 0.65 38% 25% 537 -9 0.38 64%
80
The mass in Table 18 assumes niobium, 0.3096 lbm/in3 (8570kg/m3), is used for
the entire nozzle with a throat radius of one inch. Niobium (columbium) is chosen since
it is the material for the un-cooled section of the nozzle and results in quicker
calculations. Table 18 indicated truncation of the nozzle by 75% results in a 0.4%
increase in Isp and a 8.5% decrease in mass. Why the Isp increased may be due to the
same factors causing the Isp to be inflated over the theoretical maximum. Further
reduction in length behaved as expected with a 0.2% and 1.6% decrease in Isp with a
38% and 64% decrease in weight for 50% and 25% of the total length respectively. The
results indicate truncation of the nozzle has little impact on Isp. However, as indicated in
Chapter 1, even a small change in Isp has a major impact on the capabilities of the launch
vehicle. Therefore, at this point the design will utilize the full-length nozzle. Figure 20
illustrates the full-length nozzle contour for the DEAN design.
00.5
11.5
22.5
33.5
44.5
5
0 2 4 6 8 10 12 14
L (in)
R (in
)
Figure 20. Nozzle Contour
The contour in Figure 20 comes from the contour of the TDK code presented in
Appendix C. This code matches the chamber pressure, chamber temperature, and mass
81
flow between NPSS and TDK outputs. The non-dimensional contour in this code is
dimensioned with the throat radius of 4.46 in. With an outer radius of five inches, the
throat area is 15.9 in2, which is the area input in NPSS.
4.3. Investigative Objectives
This investigation tried to meet three objectives:
1. Determine feasibility of meeting the IHPRPT Phase III orbit transfer vehicle
goals with the DEAN concept
2. Implement and improve upon a design process focused on the energy
conversion section of a rocket engine (combustion chamber, nozzle)
3. Perform detailed design analysis of the energy transfer components (cooling
jackets) making the DEAN possible
The result of the NPSS model showed that this model is able to meet the
IHPRPT Phase III goals and detailed to analysis of the energy conversion and energy
transfer components. The result is the conceptual solid model of the DEAN shown in
Figure 21.
Figure 21. The DEAN
82
The left portion of Figure 21 shows the solid model while the right portion depicts
a line drawing showing the inside of the chamber. Figure 21 illustrates the full-length
aerospike nozzle developed by TDK and the chamber designed by NPSS. The outer wall
of the chamber consists of a silicon carbide cooling jacket with an estimated mass of 5.96
lbm (2.7 kg). Around the cooling jacket is a structural jacket. The structural jacket is
sized as a pressure vessel using Equation 18:
tu
cbw F
rPt = (18)
In Equation 18, tw is the wall thickness, rc is the radius of the chamber, Ftu is the
ultimate tensile strength of the material, and Pb is the burst pressure. The bust pressure is
the chamber pressure multiplied by a factor of safety, in the case of the DEAN, the factor
of safety is 1.5. Different materials were evaluated to minimize thickness and weight.
Aluminum 7075-T6, with an Ftu of 80 ksi (555.1 MPa), was judged the best with a
thickness of 0.2 in (0.5 cm) and a mass of 13.2 lbm (6 kg). The cooled part of the nozzle
and inner wall is composed of copper resulting in a mass of about 69 lbm (31.3 kg). The
un-cooled portion of the nozzle is niobium and has a mass of 2.3 lbm (1.05 kg). The total
mass of the chamber/nozzle is estimated at 90.5 lbm (41.05 kg). Combined with the
oxygen turbo machinery at 137.3 lbm (62.3 kg) and hydrogen turbo machinery of 251.1
lbm (113.9 kg) results is an estimated total mass of 478.9 lbm (217.2 kg) for the DEAN
rocket engine concept. This results in a thrust to weight ratio of 119:1.
4.4. Summary
NPSS was used to model the DEAN. This model resulted in several significant
results. The DEAN model produces 57,231 lbf (254 kN) of thrust with an Isp of 472.7
83
sec. The hydrogen cools the inner wall of the chamber and part of the nozzle with a mass
flow of 7.55 lbm/sec (3.42 kg/sec). The cooling jacket increases the flow temperature by
464 R (257.8 K). The walls are made of copper and have a maximum temperature of
1224 R (680 K) at the throat. The oxygen cools the outer wall of the chamber with a
mass flow of 106 lbm/sec (48.1 kg/sec). The cooling jacket increases the flow
temperature by 437.4 R (243 K). The cooling channels walls consist of silicon carbide
and have a maximum temperature of 1625 R (902.8 K) at the throat. Due to the accuracy
of NPSS, it is expected that the numbers presented would closely match the results and
actual working engine.
84
5. Conclusions and Recommendations
5.1. Chapter Overview
The DEAN is a dual-expander aerospike nozzle design providing superior
performance to the current upper stage fleet. The conclusion and recommendations based
on the work done in this document are presented here.
5.2. Conclusions of Research
Based on the results presented in Chapter 4, the DEAN would meet and exceed
the IHPRPT phase III goals. Additionally, a dual-expander cycle with separate turbines
for the fuel and oxidizer flow is possible since the amount of heat captured from the
chamber is sufficient to increase the flows to the required temperature. Additionally,
though the learning curve is steep, NPSS has proven to be a powerful tool in the
development of rocket engines. TDK was easier to learn in comparison, however, its
ability to model aerospike nozzles was left wanting due to the confusing Isp result.
5.3. Significance of Research
The work contained within this document could result in significant gains in
performance for the Delta and Atlas platforms. Based on the current Centaur upper stage,
the DEAN model would result in a 28.3 s increase in Isp. This could result in as much as
a 3,806 lbm (1726 kg) increase in satellite mass to orbit. In addition, the DEAN has the
potential to save around $19 million per launch based on the savings per Isp presented in
Chapter 1.
85
5.4. Recommendations for Future Research
The work presented in this document is for the initial design concepts. Therefore,
it is recommended further research be done based on this model of the DEAN to
complete the design process. For one, the discrepancy between the Isp from the NPSS
and TDK model should be resolved. Work was done on an elements for NPSS that
allows NPSS to run TDK. At this point the element has not successfully run. Getting
this element to run should result in a improve idea of the Isp of the DEAN design and
whether truncation of the nozzle is necessary.
Even though NPSS is a powerful tool, it has some limitation. For example, it
cannot show the effects the curvature of the chamber will have on the cooling jacket and
the combustion. Therefore, a detailed analysis of the flow within the cooling jacket to
confer the properties of the flow obtained from NPSS. This would include a CFD
analysis of the fluid flow to depict the flow filed in the chamber as well as the cooling
jackets. The CFD model will result in a high fidelity model for the properties of the
chamber and their effect on the cooling jacket.
Furthermore, in this document the incorporation of the injector design only
resulted in the assumption of the pressure drop across the injector and efficiency of
combustion. A detailed investigation on the benefits of a Pintle or Coaxial injectors into
the DEAN design should be undertaken. This will determine the type of injector that
should be used on the design and a detailed design of the injectors.
The work done in this document only details a design at a single point. Therefore,
further work on the DEAN should involve on investigation on off-design conditions.
This is required to outline the ability for the model to meet throttling requirements. To
86
ensure the DEAN design is the best it can be, and optimization of the current design
should be undertaken. The optimization would determine the aspects of each component
that results in the overall best engine design. Aspects to consider are, but not limited to,
the amount of truncation of the nozzle, optimal O/F and ε, and material choices for the
walls.
Another aspect of the design to consider is the possible use of fuels other than
hydrogen. One fuel the space command has expressed interest in is the use of methane
since it is easier to store than liquid hydrogen..
5.5. Summary
In conclusion, with the incorporation of state of the art turbo machinery, and the
use of the aerospike nozzle, the proposed upper stage design of the DEAN achieves a
performance that could radically enhance the current space launch fleet. The DEAN
model resulted in 57,231 lbf (254 kN) of thrust, 14% over the goal, and an Isp of 472.7
sec, 1.9% increase over the goal. In addition, the walls of the chamber and nozzle were
adequately cooled to prevent the failure and provided the necessary energy to the fuel and
oxidizer turbine.
87
Appendix A : Lessons Learned
In the execution of the experiment detailed in this report, several lessons were
learned. Most of these lessons deal with the use of NPSS and TDK.
A.1 NPSS Lessons Learned
NPSS is a great tool, though the learning curve is steep. First, it is best not to use
the graphic-user-interface (GUI) when coding. The GUI is useful when viewing the
schematic of the code, editing, and viewing the results of the code. However, when using
it to write the code it often adds needless lines confusing more than enhancing the code.
A major revelation was the elements are defined by ‘.int’ files and were located in
a folder within NPSS. Before this revelation, the coefficients NPSS uses were confusing,
seemed unrelated to any known coefficients, and were not explained anywhere in the
documentation provided with NPSS. However, in the ‘.int’ files, the equations in each
element are defined and the coefficient values were revealed. Though the user input
coefficients were convoluted, the results of the inputs were predictable.
One drawback is NPSS only uses English units. The use of English units can
unduly confuse the problem. It would have been useful to include either SI elements, or
elements able to use either unit.
The caveat to all this is NPSS allows for the user to code there own ‘.int’ files.
Therefore, any complication can be overcome with the user creating their own elements.
However, this effort takes away from the actual design since time is spent fixing NPSS.
Additionally, learning NPSS is difficult enough, let alone creating new elements.
88
A.2 TDK Lessons Learned
TDK is a powerful program for analyzing nozzles. One useful thing would have
been graphing of the chamber as well as the nozzle. One realization in using the code
was TDK is very sensitive to the wall contour. Unlike some programs continuing to
work even if the physics are impossible, TDK aborts the code if the wall fails. This can
be frustrating when trying to design a nozzle since the contour is unknown often leaving
the researcher with a mystery as to the fix to the problem.
The ability for TDK to model aerospike nozzles is relatively unique. However,
the aerospike is a feature of the scramjet modeling capability. Some complications arise
since the scramjet is an aero-device while the aerospike is a rocket device. This creates
confusion since the scramjet burns with air while the aerospike does not use air. It is
recommended future releases of TDK have a separate aerospike function.
89
Appendix B : NPSS Code
This is the NPSS code that defines the DEAN model. This code requires the use
of modified elements developed by Tom Lavelle at NASA Glen. When opened in NPSS
these files will generate data presented in this report.
B.1 Final Model Code
//===================================================================== // Air Force Institute of Technology // 2950 Hobson Way, Bldg 641 // Wright Patterson AFB, OH 45433 // // David F. Martin II, 2Lt., USAF //===================================================================== //===================================================================== // Goals //===================================================================== real mdot = 121.25; real P_c = 1740.5; real O_F = 7; real A_throat = 15.9; real epsilon = 125; #include "Pump.int"; #include "Turb02.int"; #include "CoolingVolume.int"; real mdotO = 106; real mdotH = 15.1; real Wall_temp = 900; //===================================================================== // Chamber //===================================================================== setDefaultComposition("C_O2_H2"); RocketComb1 COMB { comp = "C_O2_H2"; radius_tc = 6.0; volume = 2075.44; Fu_I.comp = "HYDROGEN"; Fl_oxid.comp = "OXYGEN"; ThermalOutputPort Hx_zoneO1 { areaFlow = 32*PI; areaHx = 180.96; radCurv = parent.radius_tc; }; ThermalOutputPort Hx_zoneO2 { areaFlow = .COMB.Hx_zoneO1.areaFlow; areaHx = .COMB.Hx_zoneO1.areaHx;
90
radCurv = parent.radius_tc; }; ThermalOutputPort Hx_zoneO3 { areaFlow = 97.193; areaHx = .COMB.Hx_zoneO1.areaHx; radCurv = parent.radius_tc; }; ThermalOutputPort Hx_zoneO4 { areaFlow = 89.339; areaHx = .COMB.Hx_zoneO1.areaHx; radCurv = parent.radius_tc; }; ThermalOutputPort Hx_zoneO5 { areaFlow = 44.76; areaHx = 165.88; radCurv = 5.5; }; ThermalOutputPort Hx_zoneO6 { areaFlow = A_throat+3; areaHx = 3.14; radCurv = 5; }; ThermalOutputPort Hx_zoneH1 { areaFlow = 32*PI; areaHx = 60.32; radCurv = 2; }; ThermalOutputPort Hx_zoneH2 { areaFlow = .COMB.Hx_zoneH1.areaFlow; areaHx = .COMB.Hx_zoneH1.areaHx; radCurv = .COMB.Hx_zoneH1.radCurv; }; ThermalOutputPort Hx_zoneH3 { areaFlow = 97.19; areaHx = 67.86; radCurv = 2.25; }; ThermalOutputPort Hx_zoneH4 { areaFlow = 89.33; areaHx = 82.94; radCurv = 2.75; }; ThermalOutputPort Hx_zoneH5 { areaFlow = 44.76; areaHx = 120.64; radCurv = 4; }; ThermalOutputPort Hx_zoneH6 { areaFlow = A_throat+3; areaHx = 2.83; radCurv = 4.5; }; }
91
RocketNozzle NOZZ { Ath = A_throat; AR = epsilon; Ps = 0.01; ThermalOutputPort Hx_zoneH7 { areaFlow = A_throat*2; areaHx = 87.96; radCurv = 4; }; ThermalOutputPort Hx_zoneH8 { areaFlow = A_throat*4; areaHx = 54.98; radCurv = 2.5; }; } //===================================================================== // Opump //===================================================================== setDefaultComposition("OXYGEN"); Element Starter TankO {} Element Valve04 OD1 { Across = 2; K = 0.0116; } Element CoolingVolume OCV1 { UnReactedFluidInputPort Fl_I; UnReactedFluidOutputPort Fl_O; volume = 10; Aphys = 2.112; } Element Pump OP { gearRatio = 1; W = mdotO; eff = 0.773; PRdes = 103; } Element CoolingVolume OCV2 { UnReactedFluidInputPort Fl_I; UnReactedFluidOutputPort Fl_O; volume = 10; Aphys = 1.71; } //===================================================================== // Cooling Jacket1 //===================================================================== Element Valve04 DuckO1 { Across = 0.47; K = 0.0888; }
92
Element CoolingVolume CVO1 { UnReactedFluidInputPort Fl_I {OFR = 1;}; UnReactedFluidOutputPort Fl_O {OFR = 1;}; ThermalOutputPort Hx_zone1 {areaHx = 1.57;}; volume = 0.05; n_channels = 785; Aphys = 0.47; } Element Valve04 DuckO2 { Across = .CVO1.Aphys; K = 0.0879; } Element CoolingVolume CVO2 { UnReactedFluidInputPort Fl_I {OFR = 1;}; UnReactedFluidOutputPort Fl_O {OFR = 1;}; ThermalOutputPort Hx_zone1 {areaHx = 90.43;}; volume = 3.62; n_channels = .CVO1.n_channels; Aphys = 0.75; } Element Valve04 DuckO3 { Across = .CVO2.Aphys; K = 0.165; } Element CoolingVolume CVO3 { UnReactedFluidInputPort Fl_I {OFR = 1;}; UnReactedFluidOutputPort Fl_O {OFR = 1;}; ThermalOutputPort Hx_zone1 {areaHx = 105.5;}; volume = 4.22; n_channels = .CVO1.n_channels; Aphys = 0.88; } Element Valve04 DuckO4 { Across = .CVO3.Aphys; K = 0.172; } Element CoolingVolume CVO4 { UnReactedFluidInputPort Fl_I {OFR = 1;}; UnReactedFluidOutputPort Fl_O {OFR = 1;}; ThermalOutputPort Hx_zone1 {areaHx = 105.5;}; volume = 4.22; n_channels = .CVO1.n_channels; Aphys = 0.88; } Element Valve04 DuckO5 { Across = .CVO4.Aphys;
93
K = 0.127; } Element CoolingVolume CVO5 { UnReactedFluidInputPort Fl_I {OFR = 1;}; UnReactedFluidOutputPort Fl_O {OFR = 1;}; ThermalOutputPort Hx_zone1 {areaHx = 105.5;}; volume = 4.22; n_channels = .CVO1.n_channels; Aphys = 0.88; } Element Valve04 DuckO6 { Across = .CVO5.Aphys; K = 0.0968; } Element CoolingVolume CVO6 { UnReactedFluidInputPort Fl_I {OFR = 1;}; UnReactedFluidOutputPort Fl_O {OFR = 1;}; ThermalOutputPort Hx_zone1 {areaHx = 105.5;}; volume = 5.28; n_channels = .CVO1.n_channels; Aphys = 1.10; } Element Valve04 DuckO7 { Across = .CVO6.Aphys; K = 0.125; } Element Wall2 Owall1 { ThermalInputPort CoolSide1; ThermalInputPort HotSide1; } Element Wall2 Owall2 { ThermalInputPort CoolSide1; ThermalInputPort HotSide1; } Element Wall2 Owall3 { ThermalInputPort CoolSide1; ThermalInputPort HotSide1; } Element Wall2 Owall4 { ThermalInputPort CoolSide1; ThermalInputPort HotSide1; } Element Wall2 Owall5 { ThermalInputPort CoolSide1; ThermalInputPort HotSide1; } Element Wall2 Owall6 { ThermalInputPort CoolSide1; ThermalInputPort HotSide1;
94
} //===================================================================== Element CoolingVolume OCV3 { UnReactedFluidInputPort Fl_I; UnReactedFluidOutputPort Fl_O1, Fl_O2; volume = 10; Aphys = 26.42; } //===================================================================== // OT bypass //===================================================================== Element Valve04 TB1 { Across = 2; K = 194.5; } Element CoolingVolume TBCV1 { UnReactedFluidInputPort Fl_I; UnReactedFluidOutputPort Fl_O; volume = 10; Aphys = 2; } Element Valve04 TB2 { Across = 2; K = 194.5; } //===================================================================== Element Turb02 OT { Wflow = 95; Fl_O.Aphys = 14.85; eff = .949; PR = 1.8; } Element CoolingVolume OCV4 { UnReactedFluidInputPort Fl_I1 {OFR = 1;}; UnReactedFluidInputPort Fl_I2 {OFR = 1;}; UnReactedFluidOutputPort Fl_O {OFR = 1;}; volume = 10; Aphys = 14.85; } Element Valve04 OV { Across = 2; K = 0.5255; } Element Shaft OSHAFT { ShaftInputPort Sh_I1, Sh_I2; Nmech = 32000; } //=====================================================================
95
// Hpump //===================================================================== setDefaultComposition("HYDROGEN"); Element Starter TankH {} Element Valve04 HD1 { Across = 2; K = 0.0336; } Element CoolingVolume HCV1 { UnReactedFluidInputPort Fl_I; UnReactedFluidOutputPort Fl_O; volume = 10; Aphys = 2; } Element Pump HP1 { gearRatio = 1; W = mdotH; eff = .8; PRdes = 45; } Element CoolingVolume HCV2 { UnReactedFluidInputPort Fl_I; UnReactedFluidOutputPort Fl_O1, Fl_O2; volume = 10; Aphys = 2; } //===================================================================== // Bypass //===================================================================== Element Valve04 BD1 { Across = 2; K = 4; } Element CoolingVolume BCV1 { UnReactedFluidInputPort Fl_I; UnReactedFluidOutputPort Fl_O; volume = 10; Aphys = 2; } Element Valve04 BD2 { Across = 2; K = 4; } //===================================================================== Element Pump HP2 { gearRatio = 1; W = mdotH/2; eff = .83;
96
PRdes = 2; } Element CoolingVolume HCV3 { UnReactedFluidInputPort Fl_I; UnReactedFluidOutputPort Fl_O; volume = 10; Aphys = 2; } //===================================================================== // Cooling Jacket2 //===================================================================== Element Valve04 DuckH1 { Across = 0.19; K = 0.0633; } Element CoolingVolume CVH1 { UnReactedFluidInputPort Fl_I {}; UnReactedFluidOutputPort Fl_O {}; ThermalOutputPort Hx_zone1 { areaHx = 32.97; }; volume = .66; n_channels = 314; Aphys = 0.19; } Element Valve04 DuckH2 { Across = .CVH1.Aphys; K = 0.0474; } Element CoolingVolume CVH2 { UnReactedFluidInputPort Fl_I {}; UnReactedFluidOutputPort Fl_O {}; ThermalOutputPort Hx_zone1 { areaHx = 65.94; }; volume = 1.32; n_channels = 314; Aphys = 0.38; } Element Valve04 DuckH3 { Across = .CVH2.Aphys; K = 0.114; } Element CoolingVolume CVH3 { UnReactedFluidInputPort Fl_I {}; UnReactedFluidOutputPort Fl_O {}; ThermalOutputPort Hx_zone1 {
97
areaHx = 2.2; }; volume = 0.04; n_channels = 314; Aphys = 0.44; } Element Valve04 DuckH4 { Across = .CVH3.Aphys; K = 0.149; } Element CoolingVolume CVH4 { UnReactedFluidInputPort Fl_I {}; UnReactedFluidOutputPort Fl_O {}; ThermalOutputPort Hx_zone1 { areaHx = 90.43; }; volume = 1.81; n_channels = 314; Aphys = 0.38; } Element Valve04 DuckH5 { Across = .CVH4.Aphys; K = 0.0773; } Element CoolingVolume CVH5 { UnReactedFluidInputPort Fl_I {}; UnReactedFluidOutputPort Fl_O {}; ThermalOutputPort Hx_zone1 { areaHx = 52.74; }; volume = 1.58; n_channels = 314; Aphys = 0.33; } Element Valve04 DuckH6 { Across = .CVH5.Aphys; K = 0.0511; } Element CoolingVolume CVH6 { UnReactedFluidInputPort Fl_I {}; UnReactedFluidOutputPort Fl_O {}; ThermalOutputPort Hx_zone1 { areaHx = 37.68; }; volume = 1.51; n_channels = 314; Aphys = 0.31; }
98
Element Valve04 DuckH7 { Across = .CVH6.Aphys; K = 0.041; } Element CoolingVolume CVH7 { UnReactedFluidInputPort Fl_I {}; UnReactedFluidOutputPort Fl_O {}; ThermalOutputPort Hx_zone1 { areaHx = 30.14; }; volume = 1.51; n_channels = 314; Aphys = 0.31; } Element Valve04 DuckH8 { Across = .CVH7.Aphys; K = 0.0379; } Element CoolingVolume CVH8 { UnReactedFluidInputPort Fl_I {}; UnReactedFluidOutputPort Fl_O {}; ThermalOutputPort Hx_zone1 { areaHx = .CVH7.Hx_zone1.areaHx; }; volume = .CVH7.volume; n_channels = 314; Aphys = .CVH7.Aphys; } Element Valve04 DuckH9 { Across = .CVH8.Aphys; K = 0.0351; } Element Wall2 Hwall1 { ThermalInputPort CoolSide1; ThermalInputPort HotSide1; } Element Wall2 Hwall2 { ThermalInputPort CoolSide1; ThermalInputPort HotSide1; } Element Wall2 Hwall3 { ThermalInputPort CoolSide1; ThermalInputPort HotSide1; } Element Wall2 Hwall4 { ThermalInputPort CoolSide1; ThermalInputPort HotSide1; }
99
Element Wall2 Hwall5 { ThermalInputPort CoolSide1; ThermalInputPort HotSide1; } Element Wall2 Hwall6 { ThermalInputPort CoolSide1; ThermalInputPort HotSide1; } Element Wall2 Hwall7 { ThermalInputPort CoolSide1; ThermalInputPort HotSide1; } Element Wall2 Hwall8 { ThermalInputPort CoolSide1; ThermalInputPort HotSide1; } //===================================================================== Element CoolingVolume HCV4 { UnReactedFluidInputPort Fl_I; UnReactedFluidOutputPort Fl_O; volume = 10; Aphys = 2; } Element Turb02 HT { Wflow = mdotH/2; Fl_O.Aphys = 2; eff = .9; PR = 1.85; } Element CoolingVolume HCV5 { UnReactedFluidInputPort Fl_I1, Fl_I2; UnReactedFluidOutputPort Fl_O; volume = 10; Aphys = 2; } Element Valve04 HV { Across = 2; K = 2; } Element Shaft HSHAFT { ShaftInputPort Sh_I1, Sh_I2, Sh_I3; Nmech = 110000; } //===================================================================== //links //===================================================================== linkPorts("COMB.Fl_tc", "NOZZ.Fl_I", "C1"); linkPorts("TankO.Fl_O", "OD1.Fl_I", "O1"); linkPorts("OD1.Fl_O", "OCV1.Fl_I", "O2");
100
linkPorts("OCV1.Fl_O", "OP.Fl_I", "O3"); linkPorts("OP.Fl_O", "OCV2.Fl_I", "O4"); linkPorts("OCV2.Fl_O", "DuckO1.Fl_I", "O5"); linkPorts( "DuckO1.Fl_O", "CVO1.Fl_I", "O6" ); linkPorts( "CVO1.Fl_O", "DuckO2.Fl_I", "O7" ); linkPorts( "DuckO2.Fl_O", "CVO2.Fl_I", "O8" ); linkPorts( "CVO2.Fl_O", "DuckO3.Fl_I", "O9" ); linkPorts( "DuckO3.Fl_O", "CVO3.Fl_I", "O10" ); linkPorts( "CVO3.Fl_O", "DuckO4.Fl_I", "O11" ); linkPorts( "DuckO4.Fl_O", "CVO4.Fl_I", "O12" ); linkPorts( "CVO4.Fl_O", "DuckO5.Fl_I", "O13" ); linkPorts( "DuckO5.Fl_O", "CVO5.Fl_I", "O14" ); linkPorts( "CVO5.Fl_O", "DuckO6.Fl_I", "O15" ); linkPorts( "DuckO6.Fl_O", "CVO6.Fl_I", "O16" ); linkPorts( "CVO6.Fl_O", "DuckO7.Fl_I", "O17" ); linkPorts( "DuckO7.Fl_O", "OCV3.Fl_I", "O18" ); linkPorts("OCV3.Fl_O2", "TB1.Fl_I", "T1"); linkPorts("TB1.Fl_O", "TBCV1.Fl_I", "T2"); linkPorts("TBCV1.Fl_O", "TB2.Fl_I", "T3"); linkPorts("TB2.Fl_O", "OCV4.Fl_I2", "T4"); linkPorts("OCV3.Fl_O1", "OT.Fl_I", "O19"); linkPorts("OT.Fl_O", "OCV4.Fl_I1", "O20"); linkPorts("OCV4.Fl_O", "OV.Fl_I", "O21"); linkPorts("OV.Fl_O", "COMB.Fl_oxid", "O22"); linkPorts("OP.Sh_O", "OSHAFT.Sh_I1", "SO1"); linkPorts("OT.Sh_O", "OSHAFT.Sh_I2", "SO2"); linkPorts("CVO1.Hx_zone1","Owall1.CoolSide1", "W1"); linkPorts("COMB.Hx_zoneO6","Owall1.HotSide1", "W2"); linkPorts("CVO2.Hx_zone1","Owall2.CoolSide1", "W3"); linkPorts("COMB.Hx_zoneO5","Owall2.HotSide1", "W4"); linkPorts("CVO3.Hx_zone1","Owall3.CoolSide1", "W5"); linkPorts("COMB.Hx_zoneO4","Owall3.HotSide1", "W6"); linkPorts("CVO4.Hx_zone1","Owall4.CoolSide1", "W7"); linkPorts("COMB.Hx_zoneO3","Owall4.HotSide1", "W8"); linkPorts("CVO5.Hx_zone1","Owall5.CoolSide1", "W9"); linkPorts("COMB.Hx_zoneO2","Owall5.HotSide1", "W10"); linkPorts("CVO6.Hx_zone1","Owall6.CoolSide1", "W11"); linkPorts("COMB.Hx_zoneO1","Owall6.HotSide1", "W12"); linkPorts("TankH.Fl_O", "HD1.Fl_I", "F1"); linkPorts("HD1.Fl_O", "HCV1.Fl_I", "F2"); linkPorts("HCV1.Fl_O", "HP1.Fl_I", "F3"); linkPorts("HP1.Fl_O", "HCV2.Fl_I", "F4"); linkPorts("HCV2.Fl_O1", "HP2.Fl_I", "F5"); linkPorts("HCV2.Fl_O2", "BD1.Fl_I", "B1"); linkPorts("BD1.Fl_O", "BCV1.Fl_I", "B2"); linkPorts("BCV1.Fl_O", "BD2.Fl_I", "B3"); linkPorts("BD2.Fl_O", "HCV5.Fl_I2", "B4");
101
linkPorts("HP2.Fl_O", "HCV3.Fl_I", "F6"); linkPorts("HCV3.Fl_O", "DuckH1.Fl_I", "F8"); linkPorts( "DuckH1.Fl_O", "CVH1.Fl_I", "H2" ); linkPorts( "CVH1.Fl_O", "DuckH2.Fl_I", "H3" ); linkPorts( "DuckH2.Fl_O", "CVH2.Fl_I", "H4" ); linkPorts( "CVH2.Fl_O", "DuckH3.Fl_I", "H5" ); linkPorts( "DuckH3.Fl_O", "CVH3.Fl_I", "H6" ); linkPorts( "CVH3.Fl_O", "DuckH4.Fl_I", "H7" ); linkPorts( "DuckH4.Fl_O", "CVH4.Fl_I", "H8" ); linkPorts( "CVH4.Fl_O", "DuckH5.Fl_I", "H9" ); linkPorts( "DuckH5.Fl_O", "CVH5.Fl_I", "H10" ); linkPorts( "CVH5.Fl_O", "DuckH6.Fl_I", "H11" ); linkPorts( "DuckH6.Fl_O", "CVH6.Fl_I", "H12" ); linkPorts( "CVH6.Fl_O", "DuckH7.Fl_I", "H13" ); linkPorts( "DuckH7.Fl_O", "CVH7.Fl_I", "H14" ); linkPorts( "CVH7.Fl_O", "DuckH8.Fl_I", "H15" ); linkPorts( "DuckH8.Fl_O", "CVH8.Fl_I", "H16" ); linkPorts( "CVH8.Fl_O", "DuckH9.Fl_I", "H17" ); linkPorts( "DuckH9.Fl_O", "HCV4.Fl_I", "H18" ); linkPorts("HCV4.Fl_O", "HT.Fl_I", "F12"); linkPorts("HT.Fl_O", "HCV5.Fl_I1", "F13"); linkPorts("HCV5.Fl_O", "HV.Fl_I", "F15"); linkPorts("HV.Fl_O", "COMB.Fu_I", "F16"); linkPorts("HP1.Sh_O", "HSHAFT.Sh_I1", "SH1"); linkPorts("HT.Sh_O", "HSHAFT.Sh_I2", "SH2"); linkPorts("HP2.Sh_O", "HSHAFT.Sh_I3", "SH3"); linkPorts("CVH1.Hx_zone1","Hwall1.CoolSide1", "W13"); linkPorts("NOZZ.Hx_zoneH8","Hwall1.HotSide1", "W14"); linkPorts("CVH2.Hx_zone1","Hwall2.CoolSide1", "W15"); linkPorts("NOZZ.Hx_zoneH7","Hwall2.HotSide1", "W16"); linkPorts("CVH3.Hx_zone1","Hwall3.CoolSide1", "W17"); linkPorts("COMB.Hx_zoneH6","Hwall3.HotSide1", "W18"); linkPorts("CVH4.Hx_zone1","Hwall4.CoolSide1", "W19"); linkPorts("COMB.Hx_zoneH5","Hwall4.HotSide1", "W20"); linkPorts("CVH5.Hx_zone1","Hwall5.CoolSide1", "W21"); linkPorts("COMB.Hx_zoneH4","Hwall5.HotSide1", "W22"); linkPorts("CVH6.Hx_zone1","Hwall6.CoolSide1", "W23"); linkPorts("COMB.Hx_zoneH3","Hwall6.HotSide1", "W24"); linkPorts("CVH7.Hx_zone1","Hwall7.CoolSide1", "W25"); linkPorts("COMB.Hx_zoneH2","Hwall7.HotSide1", "W26"); linkPorts("CVH8.Hx_zone1","Hwall8.CoolSide1", "W27"); linkPorts("COMB.Hx_zoneH1","Hwall8.HotSide1", "W28"); //===================================================================== //Guesses //===================================================================== COMB.OFR = O_F / (1 + O_F); COMB.Tt_tc = 6550; COMB.Pt_tc = P_c; COMB.Fl_tc.W = mdot;
102
TankO.Pt = 46; TankO.Tt = 150; OD1.W = mdotO; OV.W = mdotO; DuckO1.W = mdotO; DuckO2.W = mdotO; DuckO3.W = mdotO; DuckO4.W = mdotO; DuckO5.W = mdotO; DuckO6.W = mdotO; DuckO7.W = mdotO; TB1.W = mdotO-95; TB2.W = mdotO-95; OCV1.Pt = 45; OCV1.ht = 66; OCV2.Pt = 4635; OCV2.ht = 66; CVO1.Pt = 4497; CVO1.ht = 66; CVO2.Pt = 4359; CVO2.ht = 66; CVO3.Pt = 4221; CVO3.ht = 66; CVO4.Pt = 4083; CVO4.ht = 66; CVO5.Pt = 3945; CVO5.ht = 66; CVO6.Pt = 3807; CVO6.ht = 66; OCV3.Pt = 3663; OCV3.ht = 66; TBCV1.Pt = 2849; TBCV1.ht = 66; OCV4.Pt = 2035; OCV4.ht = 66; Owall1.T = Wall_temp; Owall2.T = Wall_temp; Owall3.T = Wall_temp; Owall4.T = Wall_temp; Owall5.T = Wall_temp; Owall6.T = Wall_temp; TankH.Pt = 46; TankH.Tt = 40; HD1.W = mdotH; HV.W = mdotH; BD1.W = mdotH/2; BD2.W = mdotH/2; DuckH1.W = mdotH/2;
103
DuckH2.W = mdotH/2; DuckH3.W = mdotH/2; DuckH4.W = mdotH/2; DuckH5.W = mdotH/2; DuckH6.W = mdotH/2; DuckH7.W = mdotH/2; DuckH8.W = mdotH/2; DuckH9.W = mdotH/2; HCV1.Pt = 45; HCV1.ht = 66; HCV2.Pt = 2035; HCV2.ht = 66; BCV1.Pt = 2008; BCV1.ht = 66; HCV3.Pt = 4080; HCV3.ht = 66; CVH1.Pt = 4034; CVH1.ht = 66; CVH2.Pt = 3988; CVH2.ht = 66; CVH3.Pt = 3942; CVH3.ht = 66; CVH4.Pt = 3896; CVH4.ht = 66; CVH5.Pt = 3850; CVH5.ht = 66; CVH6.Pt = 3804; CVH6.ht = 66; CVH7.Pt = 3758; CVH7.ht = 66; CVH8.Pt = 3712; CVH8.ht = 66; HCV4.Pt = 3663; HCV4.ht = 66; HCV5.Pt = 1980; HCV5.ht = 66; Hwall1.T = Wall_temp; Hwall2.T = Wall_temp; Hwall3.T = Wall_temp; Hwall4.T = Wall_temp; Hwall5.T = Wall_temp; Hwall6.T = Wall_temp; Hwall7.T = Wall_temp; Hwall8.T = Wall_temp; //Solver will adjust these variables Independent H_valve { varName = "HP1.eff"; autoSetup = TRUE; } Independent O_valve {
104
varName = "OP.eff"; autoSetup = TRUE; } //===================================================================== // Solver //===================================================================== solver.solutionMode = "STEADY_STATE"; presolverSequence = {}; solverSequence = {"NOZZ", "TankO", "OD1", "OP", "DuckO1", "DuckO2", "DuckO3", "DuckO4", "DuckO5", "DuckO6", "DuckO7", "TB1", "TB2", "OT", "OV", "TankH", "HD1", "HP1", "BD1", "BD2", "HP2", "DuckH1", "DuckH2", "DuckH3", "DuckH4", "DuckH5", "DuckH6", "DuckH7", "DuckH8", "DuckH9", "HT", "HV", "COMB", "OCV1", "OCV2", "CVO1", "CVO2", "CVO3", "CVO4", "CVO5", "CVO6", "OCV3", "TBCV1", "OCV4", "OSHAFT", "Owall1", "Owall2", "Owall3", "Owall4", "Owall5", "Owall6", "HCV1", "HCV2", "BCV1", "HCV3", "CVH1", "CVH2", "CVH3", "CVH4", "CVH5", "CVH6", "CVH7", "CVH8", "HCV4", "HCV5", "HSHAFT","Hwall1", "Hwall2", "Hwall3", "Hwall4", "Hwall5", "Hwall6", "Hwall7", "Hwall8"} DataViewer CaseRowViewer Chamber { titleBody = ""; titleVars = {} variableList = { "COMB.Pt_tc", "COMB.Tt_tc", "COMB.OF", "NOZZ.Fg", "NOZZ.Isp", "NOZZ.W" }; } DataViewer CaseRowViewer CoolO { titleBody = ""; titleVars = {} variableList = { "CVO1.Fl_I.Tt", "CVO1.Fl_O.Tt", "CVO2.Fl_O.Tt", "CVO3.Fl_O.Tt", "CVO4.Fl_O.Tt", "CVO5.Fl_O.Tt", "CVO6.Fl_O.Tt"}; } DataViewer CaseRowViewer CoolH { titleBody = ""; titleVars = {} variableList = { "CVH1.Fl_I.Tt", "CVH1.Fl_O.Tt", "CVH2.Fl_O.Tt", "CVH3.Fl_O.Tt", "CVH4.Fl_O.Tt", "CVH5.Fl_O.Tt", "CVH6.Fl_O.Tt", "CVH7.Fl_O.Tt", "CVH8.Fl_O.Tt"}; } postsolverSequence = { "Chamber", "CoolO", "CoolH"}; autoSolverSetup(); solver.maxIterations = 5000; solver.maxJacobians = 5000; run(); Chamber.display(); CoolO.display(); CoolH.display(); cout << "Oxygen =============================="<<endl; cout << "mass Flow " << CVO1.Wavg << " lbm/sec" << endl; cout << "max temp " << Owall1.T << " R, " << Owall6.T << " R"<< endl;
105
cout << "velocity6 " << CVO6.Fl_O.V << " ft/s" << endl; cout << "velocity5 " << CVO5.Fl_O.V << " ft/s" << endl; cout << "velocity4 " << CVO4.Fl_O.V << " ft/s" << endl; cout << "Sum Q " << Owall1.Qout+Owall2.Qout+Owall3.Qout+Owall4.Qout+Owall5.Qout +Owall6.Qout << " Btu/sec" << endl; cout << "Total SA " << CVO1.Hx_zone1.areaHx+CVO2.Hx_zone1.areaHx+ CVO3.Hx_zone1.areaHx+CVO4.Hx_zone1.areaHx+CVO5.Hx_zone1.areaHx+ CVO6.Hx_zone1.areaHx << " in^2" << endl << endl; cout << "OP massFlow " << OP.W << " lbm/s, OP PR " << OP.PR << endl; cout << "ONmech " << OSHAFT.Nmech << " rpm" << endl; cout << "OT massFlow " << OT.Wflow << " lbm/s, " << " OT PR " << OT.PR << endl; cout << "OT power " << OT.pwr << " hp, OP power " << OP.pwr << " hp" << endl; cout << "Hygrogen ============================="<< endl; cout << "mass Flow " << CVH1.Wavg << " lbm/sec" << endl; cout << "max temp " << Hwall3.T << " R" << endl; cout << "velocity8 " << CVH8.Fl_O.V << " ft/s" << endl; cout << "velocity7 " << CVH7.Fl_O.V << " ft/s" << endl; cout << "velocity6 " << CVH6.Fl_O.V << " ft/s" << endl; cout << "Sum Q " << Hwall1.Qout+Hwall2.Qout+Hwall3.Qout+Hwall4.Qout+Hwall5.Qout +Hwall6.Qout+Hwall7.Qout+Hwall8.Qout << " Btu/sec" << endl; cout << "Total SA " << CVH1.Hx_zone1.areaHx+CVH2.Hx_zone1.areaHx+ CVH3.Hx_zone1.areaHx+CVH4.Hx_zone1.areaHx+CVH5.Hx_zone1.areaHx+ CVH6.Hx_zone1.areaHx+CVH7.Hx_zone1.areaHx+CVH8.Hx_zone1.areaHx << " in^2" << endl << endl; cout << "HP1 massFlow " << HP1.W << " lbm/s, HP1 PR " << HP1.PR << endl; cout << "HP2 massFlow " << HP2.W << " lbm/s, HP2 PR " << HP2.PR << endl; cout << "HNmech " << HSHAFT.Nmech << " rpm" << endl; cout << "HT massFlow " << HT.Wflow << " lbm/s " << " HT PR " << HT.PR << endl; cout << "HT pwr " << HT.pwr << " hp, HP1 pwr " << HP1.pwr << " hp, HP2 pwr " << HP2.pwr << " hp" << endl;
B.2 Included Cooling Volume Element
#ifndef __COOLINGVOLUME__ #define __COOLINGVOLUME__ //************************************************************************* // * NASA Glenn Research Center // * 21000 Brookpark Rd // * Cleveland, OH 44135 // * // ************************************************************************* //#include "/NPSS/dev/Rockets/Common/RocketIncludes.npss" #include <InterpIncludes.ncp> class CoolingVolume extends Element{ //------------------------------------------------------------ // ******* DOCUMENTATION ******* //------------------------------------------------------------ /* title = "
106
COOLING VOLUME ELEMENT ____________________ | | Any number of --> | | incoming flows | | | Cooling Volume | --> Any number of Any number of --> | | exiting flows ThermalInputPorts | | |___________________| "; */ description = isA() + " performs mass and energy storage calculations. Also allows for heat transfer."; usageNotes = " - The cooling volume element accounts for mass and energy storage at a volume location in space. - Any number of FluidInput, FluidOutput, and Thermal ports can be requested by the user at run time. These ports are then connected to the model using the usual linkPorts command. - The user may also specify and external heat transfer that is applied to the element. Note that this energy will come from outside the simulation and will break continuity of energy for the overall system. - The purpose of this element is to calculate mass and energy derivatives that are used by the solver to balance the conditions and this location in space. - The element has two default solver independents and two default solver states. - In steady-state mode the solver will vary enthalpy and pressure until the mass and energy derivatives are zero (mass in = mass out, energy in = energy out). - In transient mode, the pressure and enthalpy will be varied until the calculated internal energy and density equal the integrated energy and density. - To nullify the transient effects of this element in an otherwise transient solution, set the solution mode of the element to steady state. Ex: elementName.solutionMode = 'STEADY-STATE'. Note that this element does
107
have solutionMode as a variable. However, giving the above command will set all of the pertinent solver objects to 'STEADY-STATE'. "; //------------------------------------------------------------ // ******* SETUP VARIABLES ******** //------------------------------------------------------------ real Aphys{ IOstatus = INPUT; units = INCH2; description = "Representative cross sectional area (used to calculate statics)"; trigger = 1; } string comp{ description = "Composition of the volume"; trigger = 1; } real drhotqdt{ value = 0.0; IOstatus = OUTPUT; units = LBM_PER_FT3_SEC; description = "Time derivative of total density"; } real dutqdt{ value = 0.0; IOstatus = OUTPUT; units = BTU_PER_LBM_SEC; description = "Time derivative of total specific internal energy"; } int _Hset { value = FALSE; IOstatus = INPUT; units = NONE; description = "Used internally to determine if enthalpy has been set (User Should Not Touch)"; hide(1); } real ht{ IOstatus = OUTPUT; units = BTU_PER_LBM; description = "Total specific enthalpy"; trigger = 1; } real HtIn{ value = 0.0; IOstatus = OUTPUT; units = BTU_PER_SEC; description = "Total energy flowing in to the volume through the ports"; } real HtOut{ value = 0.0; IOstatus = OUTPUT; units = BTU_PER_SEC; description = "Total energy flowing out of the volume through the ports"; } string inportList[]; inportList { description ="list of all fluid input ports"; ptrType = "UnReactedFluidInputPort"; } int n_channels { value = 0.0; IOstatus = INPUT; units = NONE; description = "Number of channels in the cooling volume"; } string outportList[]; outportList {
108
description ="list of all fluid output ports"; ptrType = "UnReactedFluidOutputPort"; } int _Pset { value = FALSE; IOstatus = INPUT; units = NONE; description = "Used internally to determine if pressure has been set (User Should Not Touch)"; hide(1); } real Pt{ IOstatus = OUTPUT; units = PSIA; description = "Total pressure"; trigger = 1; } real Qnet{ value = 0.0; IOstatus = OUTPUT; units = BTU_PER_SEC; description = "Net heat tranfers out of the volume through thermal ports"; } real Qext{ value = 0.0; IOstatus = INPUT; units = BTU_PER_SEC; description = "External heat transfer supplied by user (note that this value breaks conservation of energy"; } real s_Qnet { value = 1.0; IOstatus = INPUT; units = NONE; description ="Heat transfer scale factor (on Q_net only)"; } real rhot{ value = 0.0; IOstatus = INPUT; units = LBM_PER_FT3; description = "Density based on total conditions"; } real ut{ value = 0.0; IOstatus = INPUT; units = BTU_PER_LBM; description = "Total specific internal energy"; } real volume{ value = 0.0; IOstatus = INPUT; units = INCH3; description = "Volume"; } real Wavg{ value = 0.0; IOstatus = OUTPUT; units = LBM_PER_SEC; description = "Average mass flowing through each port"; } real Win{ value = 0.0; IOstatus = OUTPUT; units = LBM_PER_SEC; description = "Total mass flowing into the volume"; } real Wout{ value = 0.0; IOstatus = OUTPUT; units = LBM_PER_SEC; description = "Total mass flowing out of the volume"; } //------------------------------------------------------------ // ****** SETUP PORTS, FLOW STATIONS, SOCKETS, TABLES ******** //------------------------------------------------------------
109
// FLUID PORTS UnReactedFlowStation Fl_vol{ description = "Internal station used to calculate volume properties"; } // FUEL PORTS // BLEED PORTS // THERMAL PORTS string HxPorts[]; HxPorts { description = "Array of Thermal ports"; ptrType = "ThermalOutputPort"; } // FLOW STATIONS // SOCKETS //------------------------------------------------------------ // ******* INTERNAL SOLVER SETUP ******* //------------------------------------------------------------ //------------------------------------------------------------ // ****** ADD SOLVER INDEPENDENTS & DEPENDENT ****** //------------------------------------------------------------ Independent ind_Pt { varName = "Pt"; autoSetup = TRUE; description = "Varies the volume total pressure"; } Independent ind_ht { varName = "ht"; autoSetup = TRUE; description = "Varies the volume total specific enthalpy"; indepRef = "100."; } Integrator integ_rho { stateName = "rhot"; derivativeName = "drhotqdt"; eq_rhs = "Win"; eq_lhs = "Wout"; autoSetup = TRUE; description = "Balances the volume mass storage"; } Integrator integ_U { stateName = "ut"; derivativeName = "dutqdt";
110
eq_rhs = "HtIn + Qnet + Qext"; eq_lhs = "HtOut"; autoSetup = TRUE; //eq_Ref = "100."; description = "Balances the volume energy storage"; } //------------------------------------------------------------ // ******* VARIABLE CHANGED METHODOLOGY ******* //------------------------------------------------------------ //------------------------------------------------------------ // ******* OPTION VARIABLE SETUP ******* //------------------------------------------------------------ //------------------------------------------------------------ // ******* PERFORM ENGINEERING CALCULATIONS ******* //------------------------------------------------------------ //--------------------------------------------------------------------- // calculate function //--------------------------------------------------------------------- void calculate() { int i; //------------------------------------------------------------------- // set the working area //------------------------------------------------------------------- Fl_vol.Aphys = Aphys; //------------------------------------------------------------------- // Heat Transfer Calculations //------------------------------------------------------------------- //------------------------------------------------------------------- // Determine the average flow rate //------------------------------------------------------------------- Wavg = 0.0; int iport; for ( iport = 0; iport < inportList.entries(); iport = iport + 1 ) { Wavg = Wavg + inportList[iport]->W; } for ( iport = 0; iport < outportList.entries(); iport = iport + 1 ) { Wavg = Wavg + outportList[iport]->W; } Wavg = Wavg / 2.; real Wc = Wavg; real Ac = Aphys; //------------------------------------------------------------------- // Compute the hydralic diameter //-------------------------------------------------------------------
111
real D_hyd = 2. *( Ac / PI / n_channels )**0.5; Qnet = 0; //------------------------------------------------------------------- // Loop through the thermal ports calculating the heat flux //------------------------------------------------------------------- for ( iport = 0; iport < HxPorts.entries(); iport = iport + 1 ) { //----------------------------------------------------------------- // determine the heat transfer coefficient for the heat // flow through Hx_zone1 //----------------------------------------------------------------- real hc = s_Qnet * 0.023 *( abs( Wc )/ Ac )**0.8 * Fl_vol.kt**0.6 *( Fl_vol.Cpt / Fl_vol.mut )**0.4 / D_hyd**0.2; //----------------------------------------------------------------- // Compute the heat transfer rates // Note that the area here is total for all coolant channels //----------------------------------------------------------------- real Q = hc * HxPorts[iport]->areaHx * ( HxPorts[iport]->MassTemp - Fl_vol.Tt ); //----------------------------------------------------------------- // Set the coefficients and rates in the corresponding thermal // ports connected to the hot-wall and the outer wall //----------------------------------------------------------------- HxPorts[iport]->HeatTransferCoef = hc; HxPorts[iport]->HeatTransferRate = Q; //----------------------------------------------------------------- // add the heat transfer rates to the volume overall Qhxdt //----------------------------------------------------------------- Qnet = Qnet + Q; } //------------------------------------------------------------------- // Create some working variables and initialize values for // summations //------------------------------------------------------------------- real Wt; Win = 0.0; Wout = 0.0; HtIn = 0.0; HtOut = 0.0; //------------------------------------------------------------------- // Grab the total density and internal energy from the working // station // These values are calculated by the station when Pt and ht are // set //------------------------------------------------------------------- rhot = Fl_vol.rhot; ut = Fl_vol.ut;
112
//------------------------------------------------------------------- // Sum up energy and flow over the incoming ports //------------------------------------------------------------------- // int i; for ( i=0; i < inportList.entries() ; i = i + 1 ) { Wt = inportList[i]->W; if ( Wt >= 0.0 ) { Win = Win + Wt; HtIn = HtIn + Wt * inportList[i]->ht; } else { Wout = Wout - Wt; HtOut = HtOut - Wt * ht; } } //-------------------------------------------------------------------- // Sum up energy and flow over the exiting ports //-------------------------------------------------------------------- for ( i=0; i < outportList.entries(); i = i + 1 ) { Wt = outportList[i]->W; if ( Wt >= 0.0 ) { Wout = Wout + Wt; HtOut = HtOut + Wt * ht; } else { Win = Win - Wt; HtIn = HtIn - Wt * outportList[i]->ht; } } //-------------------------------------------------------------------- // calculate the state-derivatives //-------------------------------------------------------------------- drhotqdt = ( Win - Wout )/ volume*12*12*12; dutqdt = ( HtIn + Qnet + Qext - HtOut - ut * ( Win - Wout )) / ( rhot * volume/12/12/12 ); } //-------------------------------------------------------------------- // VERIFY function (same as for basic Volume class) //-------------------------------------------------------------------- int verify() { //-------------------------------------------------------------------- // set the composition in all the stations //-------------------------------------------------------------------- int i; for ( i = 0; i < inportList.entries(); i = i + 1 ) { inportList[i]->comp = Fl_vol.comp; } for ( i = 0; i < outportList.entries(); i = i + 1 ) {
113
outportList[i]->comp = Fl_vol.comp; } //-------------------------------------------------------------------- // check to be sure the volume has a reasonable value //-------------------------------------------------------------------- if (volume <= 0.0000 ) { cerr << "verify failed because volume has not been set" << endl; return FALSE; } return TRUE; } //--------------------------------------------------------------------- // variable changed function //--------------------------------------------------------------------- void variableChanged( string name, any oldValue ){ int i; //------------------------------------------------------------------- // Put the enthalpy and pressure, which are probably set in the // solver to all the stations and ports //------------------------------------------------------------------- if ( name == "comp" ) { Fl_vol.comp = comp; //int i; for ( i = 0; i < inportList.entries(); i = i + 1 ) { inportList[i]->comp = comp; } for ( i = 0; i < outportList.entries(); i = i + 1 ) { outportList[i]->comp = comp; } } } //------------------------------------------------------------ // ******* PREPASS FUNCTION ******* //------------------------------------------------------------ void prePass() { // Set the conditions in the volume and all the ports. The // static conditions in the ports will depend on their flow // and area. int i; Fl_vol.setTotal_hP(ht, Pt); for (i=0; i<inportList.entries(); i = i + 1) { inportList[i]->setTotal_hP(ht, Pt); }
114
for (i=0; i<outportList.entries(); i = i + 1) { outportList[i]->setTotal_hP(ht, Pt); } } // ------------------------------------------------------------- // ************ POSTCREATE FUNCTION ********** // ------------------------------------------------------------- void postcreate( string name ) { //------------------------------------------------------------ // allow for the creation of ports at run time //------------------------------------------------------------ if ( name->hasInterface( "ThermalOutputPort" )) { HxPorts.append(name); } if ( name->hasInterface( "UnReactedFluidInputPort" )) { inportList.append(name); } if ( name->hasInterface( "UnReactedFluidOutputPort" )) { outportList.append( name ); } } } //dumpMaps("maps.out"); #endif
B.3 Included Pump Element
#ifndef __PUMP__ #define __PUMP__ //************************************************************************* // * NASA Glenn Research Center // * 21000 Brookpark Rd // * Cleveland, OH 44135 // * // ************************************************************************* //#include "/NPSS/dev/Rockets/Common/RocketIncludes.npss" #include <InterpIncludes.ncp> class Pump extends Element{ //------------------------------------------------------------ // ******* DOCUMENTATION ******* //------------------------------------------------------------ /* title = " PUMP ELEMENT
115
--------------------- | |---> Fl_O Fl_I - | Pump | | |---> Sh_O --------------------- | V socket: S_Map socketType: PUMP_MAP returns: head, trq "; */ description = isA() + " calculates the performance of a rocket pump element."; usageNotes = " - This element works by determining the exit conditions based on either user supplied or subelement calculated values of torque and head. - The element takes weight flow as input and determines an exit pressure. - There is a default solver independent and dependent available that will vary the weight flow until the calculated exit pressure matches the exit pressure seen at the port. - The user must supply an initial guess for the weight flow. - The inertia is kept in the mechanical port. The user should set the value directly there (Sh_O.inertia). "; //------------------------------------------------------------ // ******* SETUP VARIABLES ******** //------------------------------------------------------------ real dht { value = 0.0; IOstatus = OUTPUT; units = BTU_PER_LBM; description = "Actual specific enthalpy change"; } real dhtIdeal { value = 0.0; IOstatus = OUTPUT; units = BTU_PER_LBM; description = "Ideal specific enthalpy change"; } real eff { value = 0.0; IOstatus = OUTPUT; units = NONE; description = "Compressor efficiency"; } real gearRatio { value = 1.0; IOstatus = OUTPUT; units = NONE;
116
description = "Gear ratio on mechanical connection"; } real head { value = 0.0; IOstatus = OUTPUT; units = INCH; description = "Pump head diameter"; } real Nmech { value = 0.0; IOstatus = OUTPUT; units = RPM; description = "Rotational speed (after gear ratio applied)"; } real PR { value = 0.0; IOstatus = OUTPUT; units = NONE; description = "Pressure ratio calculated from pump characteristics"; } real PRg{ value = 0.0; IOstatus = OUTPUT; units = NONE; description = "Guess for pressure ratio calculated from fluid ports"; } real pwr { value = 0.0; IOstatus = OUTPUT; units = HORSEPOWER; description = "Compressor power"; } real trq { value = 0.0; IOstatus = OUTPUT; units = FT_LBF; description = "Compressor torque"; } real W { value = 0.0; IOstatus = INPUT; units = LBM_PER_SEC; description = "Compressor weight flow"; } real PRdes; //------------------------------------------------------------ // ****** SETUP PORTS, FLOW STATIONS, SOCKETS, TABLES ******** //------------------------------------------------------------ // FLUID PORTS UnReactedFluidInputPort Fl_I{ description = "Incoming flow"; }; UnReactedFluidOutputPort Fl_O{ description = "Exiting flow"; }; UnReactedFlowStation Fl_Otemp{ description = "Exiting flow"; }; // FUEL PORTS // BLEED PORTS // THERMAL PORTS
117
// MECHANICAL PORTS ShaftOutputPort Sh_O{ description = "Mechanical connection"; } // FLOW STATIONS UnReactedFlowStation Fl_Oi{ description = "Ideal exit conditions"; } // SOCKETS Socket S_map { allowedValues = { "trq", "head" } socketType= "PUMP_MAP"; description = "Pump performance map"; } //------------------------------------------------------------ // ****** ADD SOLVER INDEPENDENTS & DEPENDENT ****** //------------------------------------------------------------ Independent ind_W{ varName = "W"; autoSetup = TRUE; description = "Varies the weight flow"; } Dependent dep_PR{ eq_lhs = "PR"; eq_rhs = "PRdes"; autoSetup = TRUE; description = "Compares the pressure ratios calculated from the ports and the map"; } //------------------------------------------------------------ // ******* VARIABLE CHANGED METHODOLOGY ******* //------------------------------------------------------------ //------------------------------------------------------------ // ******* OPTION VARIABLE SETUP ******* //------------------------------------------------------------ Option switchDes{ allowedValues = { DESIGN, OFFDESIGN } //default is DESIGN description = "Design mode switch indicator [DESIGN / OFFDESIGN]"; trigger = 0; rewritableValues = FALSE; // Enables converter optimization. } //------------------------------------------------------------ // ******* PERFORM ENGINEERING CALCULATIONS ******* //------------------------------------------------------------ void calculate() {
118
//----------------------------------------------------------------- // Set the weight flows in the ports //----------------------------------------------------------------- Fl_I.W = W; Fl_O.W = W; //----------------------------------------------------------------- // Determine the mechanical speed from the shaft and gear ratio //----------------------------------------------------------------- Nmech = gearRatio * Sh_O.Nmech; PR = Fl_O.Pt / Fl_I.Pt; real Sout = Fl_I.s; //real PtOut = PR * PtIn; Fl_Otemp.setTotalSP( Sout, Fl_O.Pt ); real htIdealOut = Fl_Otemp.ht; //--------------------------------------------------------------- // Set the exit conditions (done to get ht right) //--------------------------------------------------------------- Fl_O.setTotal_hP( Fl_I.ht + (htIdealOut - Fl_I.ht)/eff, Fl_O.Pt ); dht = ( Fl_O.ht-Fl_I.ht ); //----------------------------------------------------------------- // Determine the power for output //----------------------------------------------------------------- pwr = -dht * W * C_BTUtoFT_LBF / C_HPtoFT_LBF_PER_SEC; //cout << "pwr is " << pwr << endl; trq = C_HP_PER_RPMtoFT_LBF * pwr / Nmech; //----------------------------------------------------------------- // Set values in the ports //----------------------------------------------------------------- Sh_O.trq = trq; }//End calculate }//End element #endif
B.4 Included Turbine Element
#ifndef __TURB02__ #define __TURB02__ //************************************************************************* // * NASA Glenn Research Center // * 21000 Brookpark Rd
119
// * Cleveland, OH 44135 // * // ************************************************************************* #include <InterpIncludes.ncp> class Turb02 extends Element{ //------------------------------------------------------------ // ******* DOCUMENTATION ******* //------------------------------------------------------------ /* title = " TURB02 ELEMENT --------------- | |---> Fl_O Fl_I --->| Turb02 | | |---> Sh_O --------------- | V socket: S_map socketType: TURB02_MAP return: eff, W "; */ description = isA() + " calculates the performance of a turbine that is being driven by a single constituent fluid"; usageNotes = " - This element works by determining the exit conditions based on either user supplied or subelement calculated values of efficiency and wieght flow. - The inertia is keep in the mechanical port. The user should set the value direclty there (Sh_O.inertia). - There are no solver independents and dependents directly associatted with this element. "; //------------------------------------------------------------ // ******* SETUP VARIABLES ******** //------------------------------------------------------------ real dht { value = 0.0; IOstatus = OUTPUT; units = BTU_PER_LBM; description = "Change in enthalpy";
120
} real dhtIdeal { value = 0.0; IOstatus = OUTPUT; units = BTU_PER_LBM; description = "Ideal enthalpy change"; } real eff { value = 0.0; IOstatus = OUTPUT; units = NONE; description = "Efficiency"; } real gearRatio { value = 1.0; IOstatus = OUTPUT; units = NONE; description = "Gear ratio on mechanical connection"; } real Nmech { value = 0.0; IOstatus = OUTPUT; units = RPM; description = "Rotational speed (after gear ratio applied)"; } real Nrad { value = 0.0; IOstatus = OUTPUT; units = RAD_PER_SEC; description = "Rotational speed in radians per sec"; } real PR { value = 0.0; IOstatus = OUTPUT; units = NONE; description = "Pressure ratio (seen from ports)"; } real pwr { value = 0.0; IOstatus = OUTPUT; units = HORSEPOWER; description = "Overall power"; } real trq { value = 0.0; IOstatus = OUTPUT; units = FT_LBF; description = "Torque on the shaft"; } real Wflow { value = 0.0; IOstatus = INPUT; units = LBM_PER_SEC; description = "Weight flow"; } //------------------------------------------------------------ // ****** SETUP PORTS, FLOW STATIONS, SOCKETS, TABLES ******** //------------------------------------------------------------ // FLUID PORTS UnReactedFluidInputPort Fl_I{ description = "Incoming flow"; }; UnReactedFluidOutputPort Fl_O{ description = "Exiting flow"; }; // FUEL PORTS
121
// BLEED PORTS // THERMAL PORTS // MECHANICAL PORTS ShaftOutputPort Sh_O{ description = "Mechanical connection"; } // FLOW STATIONS UnReactedFlowStation Fl_Oi{ description = "Ideal exit conditions"; } // SOCKETS Socket S_map { allowedValues = { "eff" , "Wflow" } description = "Turbine map"; //socketType = TURB02_MAP; } Independent ind_W{ varName = "Wflow"; //autoSetup = TRUE; description = "Varies the weight flow"; } //------------------------------------------------------------ // ******* VARIABLE CHANGED METHODOLOGY ******* //------------------------------------------------------------ //------------------------------------------------------------ // ******* OPTION VARIABLE SETUP ******* //------------------------------------------------------------ Option switchDes{ allowedValues = { DESIGN, OFFDESIGN } //default is DESIGN description = "Design mode switch indicator [DESIGN / OFFDESIGN]"; trigger = 0; rewritableValues = FALSE; // Enables converter optimization. } //------------------------------------------------------------ // ******* PERFORM ENGINEERING CALCULATIONS ******* //------------------------------------------------------------ void calculate() { //----------------------------------------------------------------- // Determine the mechanical speed from the shaft and gear ratio //----------------------------------------------------------------- Nmech = gearRatio * Sh_O.Nmech;
122
//----------------------------------------------------------------- // Switch the mechanical speed to radians per sec //----------------------------------------------------------------- Nrad = Nmech * 2 * PI / 60; //----------------------------------------------------------------- // Determine PR //----------------------------------------------------------------- PR = Fl_I.Pt / Fl_O.Pt; //------------------------------------------------------------ // determine the ideal enthalpy change //------------------------------------------------------------ Fl_Oi.copyFlow( "Fl_I" ); Fl_Oi.setTotalSP( Fl_I.s, Fl_O.Pt ); dhtIdeal = Fl_I.ht - Fl_Oi.ht; //----------------------------------------------------------------- // Calculate turbine torque //----------------------------------------------------------------- dht = dhtIdeal * eff; Fl_O.setTotal_hP( Fl_I.ht - dht, Fl_O.Pt ); trq = C_BTUtoFT_LBF * Wflow * dht / Nrad; pwr = dht * Wflow * C_BTUtoFT_LBF / C_HPtoFT_LBF_PER_SEC; //------------------------------------------------------------ // set the conditions in the shaft port //------------------------------------------------------------ Sh_O.trq = trq; //------------------------------------------------------------ // set the flows //------------------------------------------------------------ Fl_I.W = Wflow; Fl_O.W = Wflow; }//End calculate }//End element #endif
123
Appendix C : TDK Code
This is TDK input and output developed for use with the DEAN NPSS model file.
C.1 TDK Input for DEAN Upper Stage Engine
TITLE DEAN_contour100% DATA $DATA SCRJET = 1, PLANAR = F, ODE = 1, ODK=1, TDK=1, MABL = 2, IMABL = 0, MABLK = T, RSI = 2.084, RWTD = 0.05, RZERO = -1, IWALL= 2, THETA = 25, RMAX = -0.01, ZMAX = 3.2, LWALL = 1, THALW = 2.5, XLW(1) = 0, YLW(1) = -1.11, XCOWL = 0.1, ICOWL = 1, JET = T, PINF = 0.1, MINF = 16.4, GINF = 1.4, $END REACTANTS H 2. 100. -2154.L 20.27F .0709 O 2. 100. -3102.L 90.56O 1.149 NAMELISTS $ODE RKT = T, PSIA = T, XP = 1, DELH = 0, OF = T, OFSKED = 7, PSJ = 1739, TSJ = 6586, MSJ = 3.0, $END REACTIONS H + H = H2 ,M1, A = 6.4E17, N = 1.0, B = 0.0, (AR) BAULCH 72 (A) 30U H + OH = H2O ,M2, A = 8.4E21, N = 2.0, B = 0.0, (AR) BAULCH 72 (A) 10U O + O = O2 ,M3, A = 1.9E13, N = 0.0, B =-1.79,(AR) BAULCH 76 (A) 10U O + H = OH ,M7, A =3.62E18, N = 1.0, B = 0.0, (AR) JENSEN 78 (B) 30U END TBR REAX O2 + H = O + OH , A = 2.2E14, N = 0.0, B =16.8, BAULCH 72 (A) 1.5U H2 + O = H + OH , A = 1.8E10, N = -1., B = 8.9, BAULCH 72 (A) 1.5U H2 + OH = H2O + H , A = 2.2E13, N = 0.0, B =5.15, BAULCH 72 (A) 2U OH + OH = H2O + O , A = 6.3E12, N = 0.0, B =1.09, BAULCH 72 (A) 3U LAST REAX THIRD BODY REAX RATE RATIOS M1 = 25*H,4*H2,10*H2O,25*O,25*OH,1.5*O2, M2 = 12.5*H,5*H2,17*H2O,12.5*O,12.5*OH,6*O2, M3 = 12.5*H,5*H2,5*H2O,12.5*O,12.5*OH,11*O2, M7 = 12.5*H,5*H2,5*H2O,12.5*O,12.5*OH,5*O2, LAST CARD $ODK EP = 125, $END $TRANS $END $MOC $END $MABL ADBATC= 0, NTQW=2, TQW=2*2000, XTQW=-1.E6,1.E6, DXI=10E-4, NDXI=50, NYI=105, DCIMAX = .005, DXLIM(1) = .04,.75,
124
INNER = F, OFC=1, DISTRB=0, XC0=-2.16, XCE=5.13, $END $MABL ADBATC= 0, NTQW=2, TQW=2*2000, XTQW=-1.E6,1.E6, DXI=10E-4, NDXI=50, NYI=105, DCIMAX = .005, DXLIM(1) = .04,.75, INNER = T, OFC=2, DISTRB=0, XC0=-2.16, XCE=5.13, $END 0TITLE DEAN_contour100% 0DATA $DATA SCRJET = 1, PLANAR = F, ODE = 1, ODK=1, TDK=1, MABL = 2, IMABL = 0, MABLK = T, RSI = 2.084, RWTD = 0.05, RZERO = -1, IWALL= 2, THETA = 25, RMAX = -0.01, ZMAX = 3.2, LWALL = 1, THALW = 2.5, XLW(1) = 0, YLW(1) = -1.11, XCOWL = 0.1, ICOWL = 1, JET = T, PINF = 0.1, MINF = 16.4, GINF = 1.4, $END
C.2 TDK Performance Summary for the DEAN
1 TDK PERFORMANCE SUMMARY : DEAN_contour100% **********FIRST TDK SOLUTION : WITH RAMP AND COWL BOUNDARY LAYER********** REAL WALL CONTOUR 1 ZONES EXIT PLANE FIRST TDK/MABL SOLUTION THRUST CHAMBER OPERATING CONDITIONS CHAMBER PRESS [PSIA] 1739.000 CHAMBER TEMP [R] 6586.000 MIXTURE RATIO [-] 7.000000 H (OXID) [CAL/MOLE] 0.000000 H (FUEL) [CAL/MOLE] 0.000000 HCHAM (ODE) [BTU/LB] -394.6774 DELH (AVERAGE) [BTU/LB] 0.000000 DELH1 (AVE) [BTU/LB] 0.000000 THRUST CHAMBER GEOMETRY ECRAT [-] 4.000000 RI [-] 0.1000000E-01 THETAI [DEGREES] 30.00000 RWTU [-] 1.000000 RSI [INCHES] 2.084000 RWTD [-] 0.5000000E-01 NIT [-] 248.0000 THE [DEGREES] 13.06846 THETA [DEGREES] 25.00000 EP (NOZZLE) [-] 1.000010
125
ASURF [IN**2] 41.76751 EXIT FLOW PROPERTIES P (AXIS,EXIT) [PSIA] 324.6112 P (WALL,EXIT) [PSIA] 688.3699 T (WALL,EXIT) [R] 6363.542 V (WALL,EXIT) [FT/SEC] 14975.46 MA (WALL,EXIT) [-] 2.963189 ONE-DIMENSIONAL FLOW PERFORMANCE ISP (ODE) [SECONDS] 509.8608 ISP (ODK) [SECONDS] 571.7090 ISP (ODF) [SECONDS] 509.8608 CSTAR (ODE) [FT/SEC] 0.0000 CSTAR (ODK) [FT/SEC] 1462.3 CSTAR (ODF) [FT/SEC] 0.0000 TWO-DIMENSIONAL FLOW PERFORMANCE CD [-] 0.000000 CF (TDK) [-] 2.831902 CSTAR (TDK) [FT/SEC] 6302.546 THRUST (TDK) [POUNDS] 67192.85 WDOT (TDK) [LB/SEC] 121.1250 ISP (TDK) [SECONDS] 554.7396 BOUNDARY LAYER PARAMETERS DFOPT (MABLK) [POUNDS] 1176.185 DF (MABLK) [POUNDS] 1025.826 DISP (MABLK) [SECONDS] 8.469148 THETA (EXIT) [INCH] 0.8708488E-03 DEL* (EXIT) [INCH] 0.9801783E-03 DEL* (THROAT) [INCH] 0.7956905E-03 EP (REGEN) [-] 1.222427 SQDOT (REGEN) [BTU/SEC] 6711.148 SQDOT (LOSS) [BTU/SEC] 0.000000 SUM QDOT [BTU/SEC] 6711.148 DH (SUM QDOT) [BTU/LBM] 55.40678 THRUST CHAMBER PERFORMANCE THRUST (TC) [POUNDS] 66167.02 CF (TC) [-] 2.790799 WDOT (TC) [LB/SEC] 121.1250 ISP (TC) [SECONDS] 546.2704 1 TDK SCRAMJET SUMMARY : DEAN_contour100% **********FIRST TDK SOLUTION : WITH RAMP AND COWL BOUNDARY LAYER********** THRUST ON THROAT PLANE (FX BE) = 61740.66 (LBF/FT) THRUST ON EXHAUST NOZZLE (FX WALL) = 4867.923 (LBF/FT)
126
THRUST ON COWL (FX COWL) = -441.5600 (LBF/FT) TOTAL AXIAL THRUST (FX TOTAL)= 66167.02 (LBF/FT) RADIAL THRUST ON THROAT (FY BE) = 0.000000 (LBF/FT) RADIAL THRUST ON NOZZLE (FY WALL) = 0.000000 (LBF/FT) RADIAL THRUST ON COWL (FY COWL) = 0.000000 (LBF/FT) TOTAL RADIAL THRUST (FY TOTAL)= 0.000000 (LBF/FT) COWL LENGTH (XCOWL) = 0.1736667E-01 (FEET) COWL LENGTH (XCOWL) = 0.1000000 (NONDIM) XBAR WALL (XB WALL) = 0.000000 (FEET) XBAR WALL (XB WALL) = 0.000000 (NONDIM) WALL TORQUE (FY WALL)*(XBAR WALL) = 0.000000 (FT*LB/FT) YBAR WALL (YB WALL) = 0.000000 (FEET) YBAR WALL (YB WALL) = 0.000000 (NONDIM) WALL TORQUE (FX WALL)*(YBAR WALL) = 0.000000 (FT*LB/FT) XBAR BRNR (XB BRNR) = 0.000000 (FEET) XBAR BRNR (XB BRNR) = 0.000000 (NONDIM) BRNR TORQUE (FY BRNR)*(YBAR BRNR) = 0.000000 (FT*LB/FT) YBAR BRNR (YB BRNR) = 0.000000 (FEET) YBAR BRNR (YB BRNR) = 0.000000 (NONDIM) BRNR TORQUE (FX BRNR)*(YBAR BRNR) = 0.000000 (FT*LB/FT) XBAR COWL (XB COWL) = 0.000000 (FEET) XBAR COWL (XB COWL) = 0.000000 (NONDIM) COWL TORQUE (FY COWL)*(XBAR COWL) = 0.000000 (FT*LB/FT) YBAR COWL (YB COWL) = 0.000000 (FEET) YBAR COWL (YB COWL) = 0.000000 (NONDIM) COWL TORQUE (FX COWL)*(YBAR COWL) = 0.000000 (FT*LB/FT) TOTAL TORQUE ABOUT THE EXIT PLANE = 0.000000 (FT*LB/FT) PERCENTAGES OF THRUST COMPONENTS (RELATIVE TO AXIAL THRUST) THRUST ON THROAT PLANE (FX BE) = 93.31 % THRUST ON EXHAUST NOZZLE (FX WALL) = 7.36 % THRUST ON COWL (FX COWL) = -0.67 % RADIAL THRUST ON THROAT (FY BE) = 0.00 % RADIAL THRUST ON NOZZLE (FY WALL) = 0.00 % RADIAL THRUST ON COWL (FY COWL) = 0.00 % TOTAL RADIAL THRUST (FY TOTAL)= 0.00 % THRUST VECTOR ANGLE = 0.00 DEG
127
Appendix D : Isp Calculation
This appendix shows a simple calculation to demonstrate the impact or Isp hinted
to in Chapter 1. Equation 19 is the change in velocity absent drag and gravity.
bspo M
MIgu 0ln=Δ (19)
In Equation 19, Δu is the change in velocity, g0 is acceleration due to gravity, Isp
is the specific impulse, Mo is the initial mass, and Mb is the burnout mass or the mass at
the end of the thrust period. Using the gross mass of the Centaur IIA 19,073 kg as the
initial mass51, the empty mass of 2293 kg as the burn out mass51, and the stated Isp of 449
s51, result is in a Δu of:
sm
kgkgs
smu 9235
229319073ln*449*81.9 2 ==Δ
Know keeping the same Δu goal, same initial mass, but increasing the Isp by one
second yields a burn out mass of:
kgMM
kgssm
sm
bb
235419073ln*450*81.99235 2 =→=
In this analysis it is assumed the propellant mass is the initial mass minus the burn
out mass. For the lower Isp the propellant mass is 16,780 kg and for the higher Isp the
propellant mass in 16,719 kg. This less fuel in required to perform the same Δu
maneuver with an increase in Isp. The result is a savings of 61 kg (134.5 lbm). This
weight savings can mean less fuel, as shown, or larger satellites.
128
Appendix E : Wall Temperatures
This section displays the results for different materials and their affect on the wall
temperature. Table 19 shows tested materials for the hydrogen wall,
Table 19 Hydrogen Wall Materials
CVH1 CVH2 CVH3 CVH4 CVH5 CVH6 CVH7 CVH8 hH 0.00437 0.007436 0.011791 0.006182 0.003611 0.003411 0.003332 0.003317AH (in2) 54.98 87.96 2.83 120.64 82.94 67.86 60.32 60.32TH ( R ) 6586.32 6586.32 6586.32 6586.32 6586.32 6586.32 6586.32 6586.32hC 0.197015 0.131318 0.115463 0.131837 0.147505 0.154929 0.154406 0.154432AC (in2) 32.97 65.94 2.2 90.43 52.74 37.68 30.14 30.14TC ( R ) 145.391 270.699 276.61 415.765 477.206 524.952 567.469 609.732Q (Btu/s) 1439.05 3745.58 185.042 4328.38 1761.69 1349.22 1159.36 1146.54t (in) 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02Copper k (Btu/sRin) 0.005377 0.005377 0.005377 0.005377 0.005377 0.005377 0.005377 0.005377Twc ( R ) 371.6508 703.1089 980.9731 771.1716 700.7632 753.2799 813.6587 853.1702TwH ( R ) 468.999 861.4854 1224.16 904.613 779.7623 827.2277 885.1436 923.864550% Tmelt 1222.2 1222.2 1222.2 1222.2 1222.2 1222.2 1222.2 1222.2Silicon Carbide k (Btu/sRin) 0.006571 0.006571 0.006571 0.006571 0.006571 0.006571 0.006571 0.006571Twc ( R ) 372.2814 705.1291 986.1016 772.5984 701.2952 753.792 814.1962 701.2952TwH ( R ) 451.948 834.7393 1185.118 881.8026 765.9455 814.3086 872.697 765.945550% Tmelt 2790 2790 2790 2790 2790 2790 2790 2790Tungsten k (Btu/sRin) 0.002333 0.002333 0.002333 0.002333 0.002333 0.002333 0.002333 0.002333Twc ( R ) 367.121 688.5986 944.1369 760.923 696.9423 749.6009 809.7983 696.9423TwH ( R ) 591.4695 1053.593 1504.586 1068.452 879.0035 920.021 974.5421 879.003550% Tmelt 3294 3294 3294 3294 3294 3294 3294 3294Nickel k (Btu/sRin) 0.001216 0.001216 0.001216 0.001216 0.001216 0.001216 0.001216 0.001216Twc ( R ) 359.7719 665.0572 884.3742 744.2957 690.7434 743.6322 803.5352 690.7434TwH ( R ) 790.1648 1365.266 1959.546 1334.263 1040.012 1070.568 1119.582 1040.01250% Tmelt 1555.2 1555.2 1555.2 1555.2 1555.2 1555.2 1555.2 1555.2
129
In Table 19, the important values are the hot-side wall temperature (TwH) and the
50% melting temperature. As expected, the extreme TwH is for the CVH3 element,
which represents the throat. Table 19 indicates that all but the Nickel would meet the
wall temperature criteria. Moreover, Table 19 shows how the temperature of the wall is
affected by the thermal conductivity of the material (k). The oxygen side is presented in
Table 20.
Table 20 Oxygen Wall Materials
CVO1 CVO2 CVO3 CVO4 CVO5 CVO6 hH 0.011217 0.005725 0.003248 0.003003 0.002891 0.002833AH (in2) 3.14 165.88 180.96 108.96 180.96 180.96TH ( R ) 6586.32 6586.32 6586.32 6586.32 6586.32 6586.32hC 0.087317 0.062727 0.053795 0.049671 0.046118 0.03672AC (in2) 1.57 90.43 105.5 105.5 105.5 105.5TC ( R ) 179.391 303.911 379.602 452.208 531.483 616.862Q (Btu/s) 177.326 5107.5 3305.34 3019.91 2859.56 2702.71t (in) 0.02 0.02 0.02 0.02 0.02 0.02Copper k (Btu/sRin) 0.005377 0.005377 0.005377 0.005377 0.005377 0.005377Twc ( R ) 1446.081 1188.386 955.7072 806.6633 1113.574 1308.098TwH ( R ) 1656.12 1302.903 1023.642 909.7455 1172.346 1363.64750% Tmelt 1222.2 1222.2 1222.2 1222.2 1222.2 1222.2Silicon Carbide k (Btu/sRin) 0.006571 0.006571 0.006571 0.006571 0.006571 0.006571Twc ( R ) 1453.879 1191.369 956.8652 807.7637 1114.61 1309.278TwH ( R ) 1625.768 1285.086 1012.461 892.1228 1162.707 1354.73750% Tmelt 2790 2790 2790 2790 2790 2790Tungsten k (Btu/sRin) 0.002333 0.002333 0.002333 0.002333 0.002333 0.002333Twc ( R ) 1390.07 1166.962 947.3899 798.7596 1106.13 1299.628TwH ( R ) 1874.125 1430.879 1103.952 1036.323 1241.577 1427.64550% Tmelt 3294 3294 3294 3294 3294 3294
130
Table 20 cont. Nickel k (Btu/sRin) 0.001216 0.001216 0.001216 0.001216 0.001216 0.001216Twc ( R ) 1299.199 1132.204 933.8958 785.9367 1094.054 1285.886TwH ( R ) 2227.816 1638.505 1234.246 1241.681 1353.897 1531.47650% Tmelt 1555.2 1555.2 1555.2 1555.2 1555.2 1555.2Cobalt k (Btu/sRin) 0.00133 0.00133 0.00133 0.00133 0.00133 0.00133Twc ( R ) 1299.199 1132.204 933.8958 785.9367 1094.054 1285.886TwH ( R ) 2148.247 1595.122 1208.51 1202.63 1331.632 1510.43250% Tmelt 1592.1 1592.1 1592.1 1592.1 1592.1 1592.1
Table 20 shows how Twh at the throat, CVO1, excited the 50% melting point for
not only copper, but nickel and cobalt as well. Of the materials shown only silicon
carbide and tungsten meet the requirements due to their high melting points.
131
References
1 D. DeGeorge & S. Fletcher “The Integrated High Payoff Rocket Propulsion Technology Program and Tactical Missle Propulsion Status”, AFRL-PR-EN-TP-2002-214 2Futron Corporation (2006, Sept. 6) “Space Transportation Costs: Trends in Price Per Pound to Orbit 1990-2000”, Retrieved Oct. 26, 2007, from http://www.futron.com/pdf/resource_center/white_papers/Futron LaunchCostWP.pdf 3 “RL10B-2”, Pratt & Whitney Rcoketdyne, pwr_RL10B-2.pdf, 2005 4 T. Rudman & D. Del Duca, “Evolution of the Titan IV Launch Vehicle and Propulsion System”, AIAA Paper, 1998-3194. 5 B. Free, J. Owens, & F. De Poli, “THREE-BURN CENTAUR WITH ELECTRIC PROPULSION UPPER STAGE DELIVERS MAXIMUM PAYLOAD”, AIAA Paper 2002-3968. 6 R.W. Humble, G.N. Henry, & W. J. Larson, Space Propulsion Analysis and Design, 1995, McGraw-Hill Inc, New York, p.128-138. 7 P. Hill, & C. R. Peterson, Mechanics and Thermodynamics of Propulsion, second edition, 1992, Addison-Wesley Publishing Company, p. 124–130. 8 D.K Huzel., & D.H. Huang, Modern Engineering for Design of Liquid-Propellant Rocket Engines, Vol. 147, 1992, AIAA, p. 84–104. 9 S.A. Rahman, & R.J. Santoro, “A Review of Coaxial Gas/Liquid Spray Experiments and Correlations”, AIAA Paper 94-2772, 1994. 10 M.D.A. Lightfoot, S.A. Danczyk, & D.G. Talley, “Atomization in Gas-Centered Swirl-Coaxial Injectors”, DETIC Paper 50260548, 2006. 11 G.A. Dressler, & J.M. Bauer, “TRW Pintle Engine Heritage and Performance Characteristics”, AIAA Paper 2000-3871, 2000. 12 R.D. Woodward, K.L. Miller, V.G. Bazarov, G.F. Guerin, S. Pal, & R.J. Santoro, “Injector Research for Shuttle OMS Upgrade Using LOX/Ethanol Propellants”, AIAA Paper 98-3816, 1998. 13 G. Hagemann, H. Immich, T. Van Nguyen, & G. Dumnov, “Advanced Rocket Nozzles”, Journal or Propulsion and Power Vol. 14, No. 5, September-October 1998. 14 G. Hagemann, H. Immich, & M. Terhardt, “Flow Phenomena in Advanced Rocket Nozzles - Plug Nozzle”, DLR-74239/D-81663, 1998. 15 T. Ito, K. Fujii, & A. Hayashi, “Computations of the Axisymmetric Plug Nozzle Flow Fields – Flow Structures and Thrust Performance”, AIAA paper 99-3211, 1999. 16 T. Ito, & K. Fujii, “Flow Field Analysis of the Base Region of Axisymmetric Aerospike Nozzles”, AIAA Paper 2001-1051, 2001. 17 F. Nasuti, & M. Onofri, “Analysis of In-Flight Behavior of Truncated Plug Nozzles”, AIAA Paper 2000-3289, 2000. 18 T. Ito, & K. Fujii, “Numerical Analysis of the Base Bleed Effect on the Aerospike Nozzles”, AIAA Paper 2002-0512, 2002. 19 Incropera, DeWitt, Bergman, & Lavine, Fundamentals of Heat and Mass Transfer, sixth edition, 2007, John Wiley & Sons, Inc. 20 U.H. Von Glahn, “Correlation of Gas-Side Heat Transfer for Axisymmetric Rocket Engine Nozzles”, NASA TM X-1748, 1969. 21 R.L. Schacht, R.J. Quentmeyer, & W.L. Jones, “Experimental Investigation of Hot-Gas Side Heat-Transfer Rate for a Hydrogen-Oxygen Rocket”, NASA TN D-2832, 1965. 22 D.R. Boldman, J.F. Schmidt, & A.K. Gallagher, “Laminarization of a Turbulent Boundary Layer as Observed from Heat-Transfer and Boundary-Layer Measurements in Conical Nozzles”, NASA TN D-4788, 1968. 23 R. Quentmeyer, & E. Roncace, "Hot Gas Side Heat Transfer Characteristics of a Subscale Plug Nozzle Rocket Calorimeter Chamber," NASA Technical Paper 3380, 1993. 24 D. Preclik, D. Wiedmann, W. Oechslein, & J. Kretschmer, “Cryogenic Rocket Calorimeter Chamber Experiments and Heat Transfer Simulations”, AIAA Paper 1998-3440, 1998.
132
25 Jong-Gyu Kim, Kwang-Jin Lee, Seonghyeon Seo, Yeoung-Min Han, Hong-Jip Kim, & Hwan-Seok Choi, “Film Cooling Effects on Wall Heat Flux of a Liquid Propellant Combustion Chamber”, AIAA Paper 2006-5196, 2006. 26 M.H. Naragi, S. Dunn, & D. Coats, “DUAL REGENERATIVE COOLING CIRCUITS FOR LIQUID ROCKET ENGINES”, AIAA 2006-4367, 2006. 27 M.J. Chiaverini, J.A. Sauer, & S.M. Munson, “Laboratory Characterization of Vortex-Cooled Thrust Chambers for Methane/O2 and H2/O2”, AIAA 2005-4131, 2005. 28 H. Immich, & M. Caporicci, “STATUS OF THE FESTIP ROCKET PROPULSION TECHNOLOGY PROGRAMME”, AIAA Paper, 1997. 29 G. Schmidt, M. Popp, & Th. Frohlich, “Design Studies for a 10 Ton Class High Performance Expander Cycle Engine”, AIAA Paper 1998-3673, 1998. 30 R. Sorge, C. Carmicino, & A. Nocito, “Design of a Lab-Scale Cooled Two-Dimensional Plug Nozzle for Experimental Tests”, AIAA Paper 2002-4039, 2002. 31 A. Kumakawa, T. Tomita, & H. Sakamoto, “A Study of Plug-Nozzle Rocket Calorimeters”, AIAA paper 96-2583, 1996. 32 T.S. Wang, “Analysis of Linear Aerospike Plume Induced Base-Heating Physics and Fence Effect”, AIAA Paper 99-3682, 1999. 33 S. Tsutsumi, K. Yamaguchi, S. Teramoto, & T. Nagashima, “Clustering Effects on Performance and Heating of a Linear Aerospike Nozzle”, AIAA Paper 2007-122, 2007. 34 S.Peer & A. Minick, “Design and Development of an Advanced Expander Combustor”, AIAA 98-3675, 1998. 35 L. Schoenman, “4000°F Materials for Low Thrust Rocket Engines”, AIAA 93-2406, 1993. 36 A. Kumakawa, T. Onodera, M. Yoshida, M. Atsumi, & I. Igarashi, “A STUDY OF AEROSPIKE-NOZZLE ENGINES”, AIAA Paper 98-3526, 1998. 37 P.W. Gloyer, W.H. Knuth, & R.A. Crawford, “Oxygen Rich Gas Generator Design and Performance Analysis”, AIAA Paper 93-2159, 1993. 38 W. Brown, “Enhanced Heat Transfer Rocket Combustor Technology Component Hot-Fire Test Results”, AIAA Paper 90-2182, 1990. 39 J. Carlile, “An Experimental Investigation of High-Aspect-Ratio Cooling Passages”, AIAA 92-3154, 1992. 40 M.F. Wadel, & M.L. Meyer, “Validation of high aspect ratio cooling in a 89 kN (20,000 lbf) thrust combustion chamber”, AIAA 96-2584, 1996. 41 F. Neuner, D. Preclik, M. Popp, M. Funke, & H. Kluttig, “Experimental and Analytical Investigation of Local Heat Transfer in High Aspect Ratio Cooling Channels”, AIAA 98-3439, 1998. 42 R.Schuff, M. Maier, O. Sindiy, C. Ulrich, & S. Fugger, “Integrated Modeling and Analysis for a LOX/Methane Expander Cycle Engine: Focusing on Regenerative Cooling Jacket Design”, AIAA 2006-4534, 2006. 43 “NPSS User Guide”, NASA Glenn Research Center, Cleveland, OH, version 5, 30 May, 2006. 44 S.S. Dunn & D.E. Coats, “TDK’04TM Two-Dimensional Kinetics (TDK) Nozzle Performance Computer Program User’s Manual”, Software & Engineering Associates, Inc. January 2005. 45 L. Schoenman, “Final Report Oxygen Materials Compatibility Testing”, NASACR-182195, January 1989. 46 N.M. Laurendeau, “The Ignition Characteristics of Metals in Oxygen Atmosphere”, NASA Technical Report No. 851, October 1968. 47 M.A. Arguello, “The Conceptual Design of a Split Flow Liquid Hydrogen Turbopump”, AFIT Thesis, March 2008. 48 W.S. Strain, “Design of an Oxygen Turbopump for Dual Expander Cycle Rocket Engine”, AFIT Thesis, March 2008. 49“Thermophysical Properties of Fluid Systems”, National Institute of Standards and Technology, Retrieved Jan. 28, 2008, from http://webbook.nist.gov/chemistry/fluid/, 2005.
133
50 M.H. Jaskowiak, S.K. Elam, & M.R. Effinger, “Lightweight, Actively Cooled Ceramic Matric Composite Thrustcells Successfully Tested in Rocket Combustion Lab”, NASA Glenn and NASA Marshall. 51 Mark Wade, “Centaur IIA”, Encyclopedia Astronautics, Retrieved Feb. 1, 2008, from http://www.astronautix.com/stages/cenuriia.htm, 2007.
Vita
2Lt David F. Martin II graduated with the class of 2006 from Embry-Riddle
Aeronautical University (ERAU) Prescott, AZ campus. There he completed the world-
renowned aerospace engineering program, ending with a Bachelors of Science in
aerospace engineering and a Minor in math and defense studies. As a college student, he
was a member of the aeronautical honor society Sigma Gamma Tau. For the final design
project, he was a member of a group to be the first to design and fly a student built UAV
at ERAU.
2Lt Martin was commissioned through Detachment 028 at ERAU Prescott. After
completing the Air and Space Basic Course (ASBC), his first assignment was at the Air
Force Institute of Technology beginning in August 2006. Upon graduation, he will be
assigned to the 0029 Intel Squadren at Ft. Meade Army Base, Maryland.
REPORT DOCUMENTATION PAGE Form Approved OMB No. 074-0188
The public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of the collection of information, including suggestions for reducing this burden to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to an penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS. 1. REPORT DATE (DD-MM-YYYY)
14-03-2008 2. REPORT TYPE
Master’s Thesis 3. DATES COVERED (From – To)
March 2002 – March 2003 4. TITLE AND SUBTITLE COMPUTATIONAL DESIGN OF UPPERSTAGE CHAMBER, AEROSPIKE, & COOLING JACKET FOR DUAL-EXPANDER ROCKET ENGINE
5a. CONTRACT NUMBER
5b. GRANT NUMBER 5c. PROGRAM ELEMENT NUMBER
6. AUTHOR(S) Martin, David F., 2Lt, USAF
5d. PROJECT NUMBER 5e. TASK NUMBER
5f. WORK UNIT NUMBER
7. PERFORMING ORGANIZATION NAMES(S) AND ADDRESS(S) Air Force Institute of Technology Graduate School of Engineering and Management (AFIT/EN) 2950 Hobson Way, Building 640 WPAFB OH 45433-8865
8. PERFORMING ORGANIZATION REPORT NUMBER AFIT/GAE/ENY/08-M20
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) Mr. Michael Huggins Building: 8353 Room: 216B 5 Pollux Drive, Edwards CA 93524
10. SPONSOR/MONITOR’S ACRONYM(S) 11. SPONSOR/MONITOR’S REPORT NUMBER(S)
12. DISTRIBUTION/AVAILABILITY STATEMENT APPROVED FOR PUBLIC RELEASE; DISTRIBUTION UNLIMITED.
13. SUPPLEMENTARY NOTES 14. ABSTRACT To increase the performance of the current US satellite launch capability, new rocket designs must be undertaken. One concept that has been around since the 50s but yet to be utilized on a launch platform is the aerospike, or plug nozzle. The aerospike nozzle concept demonstrates globally better performance compared to a conventional bell nozzle, since the expansion of the jet is not bounded by a wall and therefore can adjust to the environment by changing the outer jet boundary. A dual-expander aerospike nozzle (DEAN) rocket concept would exceed the Integrated High Payoff Rocket Propulsion Technology initiative (IHPRPT) phase three goals. This document covers the design of the chamber and nozzle of the DEAN. The validation of the design of the DEAN are based on the model in Numerical Propulsion System Simulation (NPSS TM), added with the nozzle design from Two-Dimensional Kinematics (TDK 04TM). The result is a rocket engine that produces 57,231 lbf (254.5 kN) of thrust at an Isp of 472 s. Additionally, the oxygen wall is made of silicon carbide, with a melting point of 5580 R (3100 K), and has a maximum temperature at the throat of 1625 R (902 K). The hydrogen side is made of copper, with a melting point of 2444 R (1358 K), and has a maximum wall temperature of 1224 R (680 K) at the throat. Based on these result, future investigation into this design is merited since it has the potential to save $19 million in the cost to launch a satellite. NPSS proved to be a powerful tool in the development of rocket engines. TDK, however, was left wanting in the area of aerospike design. 15. SUBJECT TERMS Aerospike, Plug Nozzle, Regenerative Cooling, Expander Cycle, Cooling Channels, NPSS, TDK
16. SECURITY CLASSIFICATION OF:
17. LIMITATION OF ABSTRACT
UU
18. NUMBER OF PAGES
135
19a. NAME OF RESPONSIBLE PERSON Richard Branam, Maj, USAF
a. REPORT
U
b. ABSTRACT
U
c. THIS PAGE
U
19b. TELEPHONE NUMBER (Include area code) (937) 255-3636 ext. 4559 ([email protected])
Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std. Z39-18
Top Related